1 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
3 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
4 * configure: Regenerate.
6 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
8 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
10 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
12 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
15 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
17 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
20 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
22 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
24 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
27 * src/closures.c [X86_64 && __sun__ && __svr4__]
28 (FFI_MMAP_EXEC_WRIT): Define.
30 2009-12-08 David Daney <ddaney@caviumnetworks.com>
32 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
33 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
34 * testsuite/libffi.call/stret_large.c: Same.
35 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
36 * testsuite/libffi.call/stret_large2.c: Same.
37 * testsuite/libffi.call/stret_medium2.c: Same.
39 2009-12-07 David Edelsohn <edelsohn@gnu.org>
41 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
44 2009-12-05 David Edelsohn <edelsohn@gnu.org>
46 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
48 * src/powerpc/aix_closure.S: Same.
50 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
52 * Makefile.in: Regenerate.
53 * configure: Regenerate.
54 * include/Makefile.in: Regenerate.
55 * man/Makefile.in: Regenerate.
56 * testsuite/Makefile.in: Regenerate.
58 2009-12-04 David Edelsohn <edelsohn@gnu.org>
60 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
63 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
66 * src/x86/ffi64.c (classify_argument): Update from
67 gcc/config/i386/i386.c.
68 (ffi_closure_unix64_inner): Do not use the address of two consecutive
69 SSE registers directly.
70 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
71 for x86_64 linux targets.
73 2009-12-04 David Edelsohn <edelsohn@gnu.org>
75 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
76 pfr for long double split between fpr13 and stack.
78 2009-12-03 David Edelsohn <edelsohn@gnu.org>
80 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
81 fparg_count twice for long double.
83 2009-12-03 David Edelsohn <edelsohn@gnu.org>
86 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
88 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
90 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
91 Remove xfails for x86 linux targets.
93 2009-12-02 David Edelsohn <edelsohn@gnu.org>
95 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
98 2009-12-01 David Edelsohn <edelsohn@gnu.org>
100 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
101 register usage. Call ffi_prep_args directly. Add long double
102 return value support.
103 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
104 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
105 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
106 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
108 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
110 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
111 Allocate result area between params and FPRs.
113 2009-11-30 David Edelsohn <edelsohn@gnu.org>
116 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
118 * src/powerpc/aix.S: Implement AIX64 version.
119 * src/powerpc/aix_closure.S: Implement AIX64 version.
120 (ffi_closure_ASM): Use extsb, lha and displament addresses.
121 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
123 (ffi_prep_cif_machdep): Same.
125 (ffi_closure_helper_DARWIN): Same.
127 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
130 * testsuite/libffi.call/testclosure.c: New test.
132 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
134 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
135 assembly version use of ___chkstk.
137 2009-09-23 Matthias Klose <doko@ubuntu.com>
139 PR libffi/40242, PR libffi/41443
140 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
141 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
142 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
143 Change the conditionals to __SOFTFP__ || __ARM_EABI__
144 for -mfloat-abi=softfp to work.
146 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
148 PR testsuite/32843 (strikes again)
149 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
150 enable proper extension on char and short.
152 2009-09-15 David Daney <ddaney@caviumnetworks.com>
154 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
155 handling for FFI_TYPE_POINTER.
156 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
157 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
158 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
159 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
160 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
161 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
162 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
163 structure and pointer returns.
164 (ffi_closure_N32): Add handling for pointer returns.
165 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
166 calc_n32_return_struct_flags): Handle soft-float.
167 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
168 (ffi_call_N32): Declare proper argument types.
169 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
172 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
174 * configure.ac (AC_PREREQ): Bump to 2.64.
176 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
178 * Makefile.am (install-html, install-pdf): Remove.
179 * Makefile.in: Regenerate.
181 * Makefile.in: Regenerate.
182 * aclocal.m4: Regenerate.
183 * configure: Regenerate.
184 * fficonfig.h.in: Regenerate.
185 * include/Makefile.in: Regenerate.
186 * man/Makefile.in: Regenerate.
187 * testsuite/Makefile.in: Regenerate.
189 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
191 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
193 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
196 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
197 return types for X86_WIN32.
198 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
199 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
200 _ffi_closure_STDCALL): Likewise.
202 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
203 (dlmmap, dlmunmap): Also use these functions on Cygwin.
205 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
210 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
213 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
215 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
216 to "" before adding paths. (This reinstates an assignment that
217 was removed by my 2009-06-30 commit, but changes the initial
218 value from "." to "".)
220 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
223 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
224 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
227 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
229 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
230 to ld_library_path. Use add_path. Add just find_libgcc_s
231 to ld_library_path, not every libgcc multilib directory.
233 2009-06-16 Wim Lewis <wiml@hhhh.org>
235 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
236 supposed to be callee-saved.
237 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
238 return buffer for odd-size structs.
240 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
243 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
244 allow_stack_execute for Darwin.
246 2009-06-16 Andrew Haley <aph@redhat.com>
248 * configure.ac (TARGETDIR): Add missing blank lines.
249 * configure: Regenerate.
251 2009-06-16 Andrew Haley <aph@redhat.com>
253 * testsuite/libffi.call/cls_align_sint64.c,
254 testsuite/libffi.call/cls_align_uint64.c,
255 testsuite/libffi.call/cls_longdouble_va.c,
256 testsuite/libffi.call/cls_ulonglong.c,
257 testsuite/libffi.call/return_ll1.c,
258 testsuite/libffi.call/stret_medium2.c: Fix printf format
260 * testsuite/libffi.call/ffitest.h,
261 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
263 2009-06-15 Andrew Haley <aph@redhat.com>
265 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
266 * testsuite/libffi.call/err_bad_abi.c: Likewise.
268 2009-06-12 Andrew Haley <aph@redhat.com>
270 * Makefile.am: Remove info_TEXINFOS.
272 2009-06-12 Andrew Haley <aph@redhat.com>
274 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
275 testsuite/libffi.call/cls_align_uint64.c,
276 testsuite/libffi.call/cls_ulonglong.c,
277 testsuite/libffi.call/return_ll1.c,
278 testsuite/libffi.call/stret_medium2.c: Fix printf format
280 testsuite/libffi.special/unwindtest.cc: include stdint.h.
282 2009-06-11 Timothy Wall <twall@users.sf.net>
287 include/ffi_common.h,
292 src/x86/win64.S (new),
293 README: Added win64 support (mingw or MSVC)
297 testsuite/Makefile.in,
299 aclocal.m4: Regenerated
300 * ltcf-c.sh: properly escape cygwin/w32 path
301 * man/ffi_call.3: Clarify size requirements for return value.
302 * src/x86/ffi64.c: Fix filename in comment.
303 * src/x86/win32.S: Remove unused extern.
305 * testsuite/libffi.call/closure_fn0.c,
306 testsuite/libffi.call/closure_fn1.c,
307 testsuite/libffi.call/closure_fn2.c,
308 testsuite/libffi.call/closure_fn3.c,
309 testsuite/libffi.call/closure_fn4.c,
310 testsuite/libffi.call/closure_fn5.c,
311 testsuite/libffi.call/closure_fn6.c,
312 testsuite/libffi.call/closure_stdcall.c,
313 testsuite/libffi.call/cls_12byte.c,
314 testsuite/libffi.call/cls_16byte.c,
315 testsuite/libffi.call/cls_18byte.c,
316 testsuite/libffi.call/cls_19byte.c,
317 testsuite/libffi.call/cls_1_1byte.c,
318 testsuite/libffi.call/cls_20byte.c,
319 testsuite/libffi.call/cls_20byte1.c,
320 testsuite/libffi.call/cls_24byte.c,
321 testsuite/libffi.call/cls_2byte.c,
322 testsuite/libffi.call/cls_3_1byte.c,
323 testsuite/libffi.call/cls_3byte1.c,
324 testsuite/libffi.call/cls_3byte2.c,
325 testsuite/libffi.call/cls_4_1byte.c,
326 testsuite/libffi.call/cls_4byte.c,
327 testsuite/libffi.call/cls_5_1_byte.c,
328 testsuite/libffi.call/cls_5byte.c,
329 testsuite/libffi.call/cls_64byte.c,
330 testsuite/libffi.call/cls_6_1_byte.c,
331 testsuite/libffi.call/cls_6byte.c,
332 testsuite/libffi.call/cls_7_1_byte.c,
333 testsuite/libffi.call/cls_7byte.c,
334 testsuite/libffi.call/cls_8byte.c,
335 testsuite/libffi.call/cls_9byte1.c,
336 testsuite/libffi.call/cls_9byte2.c,
337 testsuite/libffi.call/cls_align_double.c,
338 testsuite/libffi.call/cls_align_float.c,
339 testsuite/libffi.call/cls_align_longdouble.c,
340 testsuite/libffi.call/cls_align_longdouble_split.c,
341 testsuite/libffi.call/cls_align_longdouble_split2.c,
342 testsuite/libffi.call/cls_align_pointer.c,
343 testsuite/libffi.call/cls_align_sint16.c,
344 testsuite/libffi.call/cls_align_sint32.c,
345 testsuite/libffi.call/cls_align_sint64.c,
346 testsuite/libffi.call/cls_align_uint16.c,
347 testsuite/libffi.call/cls_align_uint32.c,
348 testsuite/libffi.call/cls_align_uint64.c,
349 testsuite/libffi.call/cls_dbls_struct.c,
350 testsuite/libffi.call/cls_double.c,
351 testsuite/libffi.call/cls_double_va.c,
352 testsuite/libffi.call/cls_float.c,
353 testsuite/libffi.call/cls_longdouble.c,
354 testsuite/libffi.call/cls_longdouble_va.c,
355 testsuite/libffi.call/cls_multi_schar.c,
356 testsuite/libffi.call/cls_multi_sshort.c,
357 testsuite/libffi.call/cls_multi_sshortchar.c,
358 testsuite/libffi.call/cls_multi_uchar.c,
359 testsuite/libffi.call/cls_multi_ushort.c,
360 testsuite/libffi.call/cls_multi_ushortchar.c,
361 testsuite/libffi.call/cls_pointer.c,
362 testsuite/libffi.call/cls_pointer_stack.c,
363 testsuite/libffi.call/cls_schar.c,
364 testsuite/libffi.call/cls_sint.c,
365 testsuite/libffi.call/cls_sshort.c,
366 testsuite/libffi.call/cls_uchar.c,
367 testsuite/libffi.call/cls_uint.c,
368 testsuite/libffi.call/cls_ulonglong.c,
369 testsuite/libffi.call/cls_ushort.c,
370 testsuite/libffi.call/err_bad_abi.c,
371 testsuite/libffi.call/err_bad_typedef.c,
372 testsuite/libffi.call/float2.c,
373 testsuite/libffi.call/huge_struct.c,
374 testsuite/libffi.call/nested_struct.c,
375 testsuite/libffi.call/nested_struct1.c,
376 testsuite/libffi.call/nested_struct10.c,
377 testsuite/libffi.call/nested_struct2.c,
378 testsuite/libffi.call/nested_struct3.c,
379 testsuite/libffi.call/nested_struct4.c,
380 testsuite/libffi.call/nested_struct5.c,
381 testsuite/libffi.call/nested_struct6.c,
382 testsuite/libffi.call/nested_struct7.c,
383 testsuite/libffi.call/nested_struct8.c,
384 testsuite/libffi.call/nested_struct9.c,
385 testsuite/libffi.call/problem1.c,
386 testsuite/libffi.call/return_ldl.c,
387 testsuite/libffi.call/return_ll1.c,
388 testsuite/libffi.call/stret_large.c,
389 testsuite/libffi.call/stret_large2.c,
390 testsuite/libffi.call/stret_medium.c,
391 testsuite/libffi.call/stret_medium2.c,
392 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
393 of checking for MMAP. Use intptr_t instead of long casts.
395 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
397 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
398 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
399 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
401 2009-06-09 Andrew Haley <aph@redhat.com>
403 * src/x86/freebsd.S: Add missing file.
405 2009-06-08 Andrew Haley <aph@redhat.com>
407 Import from libffi 3.0.8:
409 * doc/libffi.texi: New file.
410 * doc/libffi.info: Likewise.
411 * doc/stamp-vti: Likewise.
412 * man/Makefile.am: New file.
413 * man/ffi_call.3: New file.
415 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
417 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
419 * configure.ac: Bump version to 3.0.8.
420 parisc*-*-linux*: Add.
421 i386-*-freebsd* | i386-*-openbsd*: Add.
422 powerpc-*-beos*: Add.
423 AM_CONDITIONAL X86_FREEBSD: Add.
424 AC_CONFIG_FILES: Add man/Makefile.
426 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
428 2009-06-08 Andrew Haley <aph@redhat.com>
430 * README: Import from libffi 3.0.8.
432 2009-06-08 Andrew Haley <aph@redhat.com>
434 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
435 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
436 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
437 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
439 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
440 * testsuite/libffi.call/stret_medium.c: Likewise.
441 * testsuite/libffi.call/stret_large2.c: Likewise.
442 * testsuite/libffi.call/stret_large.c: Likewise.
444 2008-12-26 Timothy Wall <twall@users.sf.net>
446 * testsuite/libffi.call/cls_longdouble.c,
447 testsuite/libffi.call/cls_longdouble_va.c,
448 testsuite/libffi.call/cls_align_longdouble.c,
449 testsuite/libffi.call/cls_align_longdouble_split.c,
450 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
451 failures on x86_64 cygwin/mingw.
453 2008-12-22 Timothy Wall <twall@users.sf.net>
455 * testsuite/libffi.call/closure_fn0.c,
456 testsuite/libffi.call/closure_fn1.c,
457 testsuite/libffi.call/closure_fn2.c,
458 testsuite/libffi.call/closure_fn3.c,
459 testsuite/libffi.call/closure_fn4.c,
460 testsuite/libffi.call/closure_fn5.c,
461 testsuite/libffi.call/closure_fn6.c,
462 testsuite/libffi.call/closure_loc_fn0.c,
463 testsuite/libffi.call/closure_stdcall.c,
464 testsuite/libffi.call/cls_align_pointer.c,
465 testsuite/libffi.call/cls_pointer.c,
466 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
467 pointer to integer (intptr_t).
468 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
470 2008-07-24 Anthony Green <green@redhat.com>
472 * testsuite/libffi.call/cls_dbls_struct.c,
473 testsuite/libffi.call/cls_double_va.c,
474 testsuite/libffi.call/cls_longdouble.c,
475 testsuite/libffi.call/cls_longdouble_va.c,
476 testsuite/libffi.call/cls_pointer.c,
477 testsuite/libffi.call/cls_pointer_stack.c,
478 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
481 2008-03-04 Anthony Green <green@redhat.com>
485 * testsuite/libffi.call/cls_align_longdouble_split2.c
486 testsuite/libffi.call/cls_align_longdouble_split.c
487 testsuite/libffi.call/cls_dbls_struct.c
488 testsuite/libffi.call/cls_double_va.c
489 testsuite/libffi.call/cls_longdouble.c
490 testsuite/libffi.call/cls_longdouble_va.c
491 testsuite/libffi.call/cls_pointer.c
492 testsuite/libffi.call/cls_pointer_stack.c
493 testsuite/libffi.call/err_bad_abi.c
494 testsuite/libffi.call/err_bad_typedef.c
495 testsuite/libffi.call/stret_large2.c
496 testsuite/libffi.call/stret_large.c
497 testsuite/libffi.call/stret_medium2.c
498 testsuite/libffi.call/stret_medium.c: New tests from Apple.
500 2009-06-05 Andrew Haley <aph@redhat.com>
502 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
505 2009-06-04 Andrew Haley <aph@redhat.com>
507 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
510 2008-02-26 Anthony Green <green@redhat.com>
511 Thomas Heller <theller@ctypes.org>
513 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
516 2008-02-03 Timothy Wall <twall@users.sf.net>
518 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
519 offset based on code pointer, not data pointer.
521 2008-01-31 Timothy Wall <twall@users.sf.net>
523 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
525 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
527 * src/x86/win32.S: Add assembly for stdcall closure.
528 * src/x86/ffi.c: Initialize stdcall closure trampoline.
530 2009-06-04 Andrew Haley <aph@redhat.com>
532 * include/ffi.h.in: Change void (*)() to void (*)(void).
533 * src/x86/ffi.c: Likewise.
535 2009-06-04 Andrew Haley <aph@redhat.com>
537 * src/powerpc/ppc_closure.S: Insert licence header.
538 * src/powerpc/linux64_closure.S: Likewise.
539 * src/m68k/sysv.S: Likewise.
541 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
542 * src/powerpc/ffi.c: Likewise.
543 * src/powerpc/ffi_darwin.c: Likewise.
544 * src/m32r/ffi.c: Likewise.
545 * src/sh64/ffi.c: Likewise.
546 * src/x86/ffi64.c: Likewise.
547 * src/alpha/ffi.c: Likewise.
548 * src/alpha/osf.S: Likewise.
549 * src/frv/ffi.c: Likewise.
550 * src/s390/ffi.c: Likewise.
551 * src/pa/ffi.c: Likewise.
552 * src/pa/hpux32.S: Likewise.
553 * src/ia64/unix.S: Likewise.
554 * src/ia64/ffi.c: Likewise.
555 * src/sparc/ffi.c: Likewise.
556 * src/mips/ffi.c: Likewise.
557 * src/sh/ffi.c: Likewise.
559 2008-02-15 David Daney <ddaney@avtrex.com>
561 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
562 Define (conditionally), and use it to include cachectl.h.
563 (ffi_prep_closure_loc): Fix cache flushing.
564 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
566 2009-06-04 Andrew Haley <aph@redhat.com>
572 src/powerpc/ffitarget.h,
574 src/sh64/ffitarget.h,
585 src/alpha/ffitarget.h,
588 src/m68k/ffitarget.h,
591 src/s390/ffitarget.h,
593 src/cris/ffitarget.h,
598 src/ia64/ffitarget.h,
601 src/ia64/ia64_flags.h,
605 src/sparc/ffitarget.h,
608 src/mips/ffitarget.h,
615 src/sh/sysv.S: Update license text.
617 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
619 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
620 (.eh_frame): Add FDE for it.
622 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
624 * configure.ac: Also check if assembler supports pc-relative
625 relocs on X86_WIN32 targets.
626 * configure: Regenerate.
627 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
628 (_ffi_call_SYSV): Add missing function type symbol .def and
629 add EH markup labels.
630 (_ffi_call_STDCALL): Likewise.
631 (_ffi_closure_SYSV): Likewise.
632 (_ffi_closure_raw_SYSV): Likewise.
633 (.eh_frame): Add hand-crafted EH data.
635 2009-04-09 Jakub Jelinek <jakub@redhat.com>
637 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
638 version 3 of the GNU General Public License and to point readers
639 at the COPYING3 file and the FSF's license web page.
640 * testsuite/libffi.call/call.exp: Likewise.
641 * testsuite/libffi.special/special.exp: Likewise.
643 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
645 * configure: Regenerate.
647 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
650 * configure.ac (HAVE_AS_X86_PCREL): New test.
651 * configure: Regenerate.
652 * fficonfig.h.in: Regenerate.
653 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
654 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
655 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
656 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
657 * src/x86/unix64.S (.Lstore_table): Move to .text section.
658 (.Lload_table): Likewise.
659 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
661 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
663 * configure: Regenerate.
665 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
667 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
668 signed/unsigned int8/16 return values.
669 * src/sparc/v8.S (ffi_call_v8): Likewise.
670 (ffi_closure_v8): Likewise.
672 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
673 Steve Ellcey <sje@cup.hp.com>
675 * configure: Regenerate for new libtool.
676 * Makefile.in: Ditto.
677 * include/Makefile.in: Ditto.
680 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
682 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
683 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
684 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
685 Adjust copyright notice.
686 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
687 register or two register to use for FFI_SYSV structs.
688 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
689 (ffi_closure_helper_SYSV): Fix the return type for
690 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
691 Adjust copyright notice.
693 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
695 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
698 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
700 * configure: Regenerate.
701 * include/Makefile.in: Regenerate.
702 * testsuite/Makefile.in: Regenerate.
704 2008-06-07 Joseph Myers <joseph@codesourcery.com>
706 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
707 powerpc-*-beos*): Remove.
708 * configure: Regenerate.
710 2008-05-09 Julian Brown <julian@codesourcery.com>
712 * Makefile.am (LTLDFLAGS): New.
713 (libffi_la_LDFLAGS): Use above.
714 * Makefile.in: Regenerate.
716 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
719 * aclocal.m4: Regenerate.
720 * configure: Regenerate.
722 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
724 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
725 * src/sh64/sysv.S: Likewise.
727 2008-03-26 Daniel Jacobowitz <dan@debian.org>
729 * src/arm/sysv.S: Fix ARM comment marker.
731 2008-03-26 Jakub Jelinek <jakub@redhat.com>
733 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
734 * src/s390/sysv.S: Likewise.
735 * src/powerpc/ppc_closure.S: Likewise.
736 * src/powerpc/sysv.S: Likewise.
737 * src/x86/unix64.S: Likewise.
738 * src/x86/sysv.S: Likewise.
739 * src/sparc/v8.S: Likewise.
740 * src/sparc/v9.S: Likewise.
741 * src/m68k/sysv.S: Likewise.
742 * src/arm/sysv.S: Likewise.
744 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
746 * aclocal.m4: Regenerate.
747 * configure: Likewise.
748 * Makefile.in: Likewise.
749 * include/Makefile.in: Likewise.
750 * testsuite/Makefile.in: Likewise.
752 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
753 Andreas Tobler <a.tobler@schweiz.org>
755 * configure.ac: Add amd64-*-freebsd* target.
756 * configure: Regenerate.
758 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
761 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
764 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
767 2008-01-24 David Edelsohn <edelsohn@gnu.org>
769 * configure: Regenerate.
771 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
773 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
775 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
778 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
779 signed/unsigned int8/16 for X86_DARWIN.
780 Updated copyright info.
781 Handle one and two byte structs with special cif->flags.
782 * src/x86/ffitarget.h: Add special types for one and two byte structs.
783 Updated copyright info.
784 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
786 Remove code to pop args from the stack after call.
787 Special-case signed/unsigned for int8/16, one and two byte structs.
788 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
789 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
791 Updated copyright info.
793 2007-12-08 David Daney <ddaney@avtrex.com>
795 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
796 SUBU, add with ADDU and use smaller code sequences.
798 2007-12-07 David Daney <ddaney@avtrex.com>
800 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
803 2007-12-06 David Daney <ddaney@avtrex.com>
805 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
807 (ffi_java_raw): New typedef.
808 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
809 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
811 (ffi_java_raw_closure) : Same.
812 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
814 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
816 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
817 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
818 sizeof(ffi_java_raw) for alignment calculations.
819 (ffi_java_ptrarray_to_raw): Same.
820 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
821 if FFI_SIZEOF_JAVA_RAW == 4.
822 (ffi_java_raw_to_rvalue): Same.
823 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
824 (ffi_java_translate_args): Same.
825 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
827 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
829 2007-12-06 David Daney <ddaney@avtrex.com>
831 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
834 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
837 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
838 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
839 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
840 set the NUM_FPR_ARG_REGISTERS according to.
841 Add support for potential soft-float support under hard-float
843 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
844 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
845 to the FFI_LINUX_SOFT_FLOAT ABI.
846 (ffi_prep_cif_machdep): Likewise.
847 (ffi_closure_helper_SYSV): Likewise.
848 * src/powerpc/ppc_closure.S: Make sure not to store float/double
849 on archs where __NO_FPRS__ is true.
850 Add FFI_TYPE_UINT128 support.
851 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
852 Adjust copyright notice.
854 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
856 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
857 * include/ffi_common.h: ... here.
860 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
862 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
863 * src/powerpc/linux64.S: Likewise.
864 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
865 * testsuite/libffi.call/return_ldl.c: New test.
867 2007-09-04 <aph@redhat.com>
869 * src/arm/sysv.S (UNWIND): New.
870 (Whole file): Conditionally compile unwinder directives.
871 * src/arm/sysv.S: Add unwinder directives.
873 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
874 Only treat r0 as a struct address if we're actually returning a
876 Only copy the bytes that are actually within a struct.
877 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
878 is returned in r0, not passed by address.
879 (ffi_call): Allocate a word-sized temporary for the case where
880 a composite is returned in r0.
881 (ffi_prep_incoming_args_SYSV): Align as necessary.
883 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
885 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
886 directly using the sys_cacheflush syscall.
888 2007-07-27 Andrew Haley <aph@redhat.com>
890 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
892 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
894 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
895 * configure.ac: Likewise.
896 * Makefile.in: Regenerate.
897 * include/Makefile.in: Likewise.
898 * testsuite/Makefile.in: Likewise.
899 * configure: Likewise.
901 2007-08-24 David Daney <ddaney@avtrex.com>
903 * testsuite/libffi.call/return_sl.c: New test.
905 2007-08-10 David Daney <ddaney@avtrex.com>
907 * testsuite/libffi.call/cls_multi_ushort.c,
908 testsuite/libffi.call/cls_align_uint16.c,
909 testsuite/libffi.call/nested_struct1.c,
910 testsuite/libffi.call/nested_struct3.c,
911 testsuite/libffi.call/cls_7_1_byte.c,
912 testsuite/libffi.call/nested_struct5.c,
913 testsuite/libffi.call/cls_double.c,
914 testsuite/libffi.call/nested_struct7.c,
915 testsuite/libffi.call/cls_sint.c,
916 testsuite/libffi.call/nested_struct9.c,
917 testsuite/libffi.call/cls_20byte1.c,
918 testsuite/libffi.call/cls_multi_sshortchar.c,
919 testsuite/libffi.call/cls_align_sint64.c,
920 testsuite/libffi.call/cls_3byte2.c,
921 testsuite/libffi.call/cls_multi_schar.c,
922 testsuite/libffi.call/cls_multi_uchar.c,
923 testsuite/libffi.call/cls_19byte.c,
924 testsuite/libffi.call/cls_9byte1.c,
925 testsuite/libffi.call/cls_align_float.c,
926 testsuite/libffi.call/closure_fn1.c,
927 testsuite/libffi.call/problem1.c,
928 testsuite/libffi.call/closure_fn3.c,
929 testsuite/libffi.call/cls_sshort.c,
930 testsuite/libffi.call/closure_fn5.c,
931 testsuite/libffi.call/cls_align_double.c,
932 testsuite/libffi.call/nested_struct.c,
933 testsuite/libffi.call/cls_2byte.c,
934 testsuite/libffi.call/nested_struct10.c,
935 testsuite/libffi.call/cls_4byte.c,
936 testsuite/libffi.call/cls_6byte.c,
937 testsuite/libffi.call/cls_8byte.c,
938 testsuite/libffi.call/cls_multi_sshort.c,
939 testsuite/libffi.call/cls_align_sint16.c,
940 testsuite/libffi.call/cls_align_uint32.c,
941 testsuite/libffi.call/cls_20byte.c,
942 testsuite/libffi.call/cls_float.c,
943 testsuite/libffi.call/nested_struct2.c,
944 testsuite/libffi.call/cls_5_1_byte.c,
945 testsuite/libffi.call/nested_struct4.c,
946 testsuite/libffi.call/cls_24byte.c,
947 testsuite/libffi.call/nested_struct6.c,
948 testsuite/libffi.call/cls_64byte.c,
949 testsuite/libffi.call/nested_struct8.c,
950 testsuite/libffi.call/cls_uint.c,
951 testsuite/libffi.call/cls_multi_ushortchar.c,
952 testsuite/libffi.call/cls_schar.c,
953 testsuite/libffi.call/cls_uchar.c,
954 testsuite/libffi.call/cls_align_uint64.c,
955 testsuite/libffi.call/cls_ulonglong.c,
956 testsuite/libffi.call/cls_align_longdouble.c,
957 testsuite/libffi.call/cls_1_1byte.c,
958 testsuite/libffi.call/cls_12byte.c,
959 testsuite/libffi.call/cls_3_1byte.c,
960 testsuite/libffi.call/cls_3byte1.c,
961 testsuite/libffi.call/cls_4_1byte.c,
962 testsuite/libffi.call/cls_6_1_byte.c,
963 testsuite/libffi.call/cls_16byte.c,
964 testsuite/libffi.call/cls_18byte.c,
965 testsuite/libffi.call/closure_fn0.c,
966 testsuite/libffi.call/cls_9byte2.c,
967 testsuite/libffi.call/closure_fn2.c,
968 testsuite/libffi.call/closure_fn4.c,
969 testsuite/libffi.call/cls_ushort.c,
970 testsuite/libffi.call/closure_fn6.c,
971 testsuite/libffi.call/cls_5byte.c,
972 testsuite/libffi.call/cls_align_pointer.c,
973 testsuite/libffi.call/cls_7byte.c,
974 testsuite/libffi.call/cls_align_sint32.c,
975 testsuite/libffi.special/unwindtest_ffi_call.cc,
976 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
978 2007-08-10 David Daney <ddaney@avtrex.com>
981 * configure.ac: Don't treat mips64 as a special case.
982 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
983 * configure: Regenerate
984 * Makefile.in: Ditto.
985 * fficonfig.h.in: Ditto.
986 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
987 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
988 (FFI_DEFAULT_ABI): Set for n64 case.
989 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
990 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
991 (ffi_closure_N32): New function.
992 (.eh_frame): New section
993 * src/mips/o32.S: Clean up comments.
994 (ffi_closure_O32): Pass ffi_closure parameter in $12.
995 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
996 _MIPS_SIM == _ABIN32 throughout.
997 (FFI_MIPS_STOP_HERE): New, use in place of
999 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1000 to support n32/n64 ABIs.
1001 (calc_n32_struct_flags): Rewrite.
1002 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1003 position of flag bits.
1004 (ffi_prep_cif_machdep): Rewrite n32 portion.
1005 (ffi_call): Enable for n64. Add special handling for small structure
1007 (ffi_prep_closure_loc): Add n32 and n64 support.
1008 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1009 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1011 2007-08-08 David Daney <ddaney@avtrex.com>
1013 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1014 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1016 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1017 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1018 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1019 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1020 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1021 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1022 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1023 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1024 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1025 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1026 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1027 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1028 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1029 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1030 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1031 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1032 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1033 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1034 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1035 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1036 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1037 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1038 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1039 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1040 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1041 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1042 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1043 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1044 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1045 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1046 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1048 2007-08-07 Andrew Haley <aph@redhat.com>
1050 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1053 2007-08-06 Andrew Haley <aph@redhat.com>
1056 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1057 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1060 2007-08-02 David Daney <ddaney@avtrex.com>
1062 * testsuite/libffi.call/return_ul.c (main): Define return type as
1063 ffi_arg. Use proper printf conversion specifier.
1065 2007-07-30 Andrew Haley <aph@redhat.com>
1068 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1069 signed/unsigned int8/16.
1070 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1072 Remove code to pop args from the stack after call.
1073 Special-case signed/unsigned int8/16.
1074 * testsuite/libffi.call/return_sc.c (main): Revert.
1076 2007-07-26 Richard Guenther <rguenther@suse.de>
1079 * testsuite/libffi.call/return_sc.c (main): Verify call
1080 result as signed char, not ffi_arg.
1082 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1084 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1085 * configure: Regenerate.
1087 2007-07-11 David Daney <ddaney@avtrex.com>
1089 * src/mips/ffi.c: Don't include sys/cachectl.h.
1090 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1093 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1095 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1096 from (ffi_prep_closure): ... this.
1097 (FFI_INIT_TRAMPOLINE): Adjust.
1099 2005-12-31 Phil Blundell <pb@reciva.com>
1101 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1102 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1103 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1104 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1105 (FFI_CLOSURES): Enable closure support.
1107 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1109 * testsuite/libffi.call/cls_multi_ushort.c,
1110 testsuite/libffi.call/cls_align_uint16.c,
1111 testsuite/libffi.call/nested_struct1.c,
1112 testsuite/libffi.call/nested_struct3.c,
1113 testsuite/libffi.call/cls_7_1_byte.c,
1114 testsuite/libffi.call/cls_double.c,
1115 testsuite/libffi.call/nested_struct5.c,
1116 testsuite/libffi.call/nested_struct7.c,
1117 testsuite/libffi.call/cls_sint.c,
1118 testsuite/libffi.call/nested_struct9.c,
1119 testsuite/libffi.call/cls_20byte1.c,
1120 testsuite/libffi.call/cls_multi_sshortchar.c,
1121 testsuite/libffi.call/cls_align_sint64.c,
1122 testsuite/libffi.call/cls_3byte2.c,
1123 testsuite/libffi.call/cls_multi_schar.c,
1124 testsuite/libffi.call/cls_multi_uchar.c,
1125 testsuite/libffi.call/cls_19byte.c,
1126 testsuite/libffi.call/cls_9byte1.c,
1127 testsuite/libffi.call/cls_align_float.c,
1128 testsuite/libffi.call/closure_fn1.c,
1129 testsuite/libffi.call/problem1.c,
1130 testsuite/libffi.call/closure_fn3.c,
1131 testsuite/libffi.call/cls_sshort.c,
1132 testsuite/libffi.call/closure_fn5.c,
1133 testsuite/libffi.call/cls_align_double.c,
1134 testsuite/libffi.call/cls_2byte.c,
1135 testsuite/libffi.call/nested_struct.c,
1136 testsuite/libffi.call/nested_struct10.c,
1137 testsuite/libffi.call/cls_4byte.c,
1138 testsuite/libffi.call/cls_6byte.c,
1139 testsuite/libffi.call/cls_8byte.c,
1140 testsuite/libffi.call/cls_multi_sshort.c,
1141 testsuite/libffi.call/cls_align_uint32.c,
1142 testsuite/libffi.call/cls_align_sint16.c,
1143 testsuite/libffi.call/cls_float.c,
1144 testsuite/libffi.call/cls_20byte.c,
1145 testsuite/libffi.call/cls_5_1_byte.c,
1146 testsuite/libffi.call/nested_struct2.c,
1147 testsuite/libffi.call/cls_24byte.c,
1148 testsuite/libffi.call/nested_struct4.c,
1149 testsuite/libffi.call/nested_struct6.c,
1150 testsuite/libffi.call/cls_64byte.c,
1151 testsuite/libffi.call/nested_struct8.c,
1152 testsuite/libffi.call/cls_uint.c,
1153 testsuite/libffi.call/cls_multi_ushortchar.c,
1154 testsuite/libffi.call/cls_schar.c,
1155 testsuite/libffi.call/cls_uchar.c,
1156 testsuite/libffi.call/cls_align_uint64.c,
1157 testsuite/libffi.call/cls_ulonglong.c,
1158 testsuite/libffi.call/cls_align_longdouble.c,
1159 testsuite/libffi.call/cls_1_1byte.c,
1160 testsuite/libffi.call/cls_12byte.c,
1161 testsuite/libffi.call/cls_3_1byte.c,
1162 testsuite/libffi.call/cls_3byte1.c,
1163 testsuite/libffi.call/cls_4_1byte.c,
1164 testsuite/libffi.call/cls_6_1_byte.c,
1165 testsuite/libffi.call/cls_16byte.c,
1166 testsuite/libffi.call/cls_18byte.c,
1167 testsuite/libffi.call/closure_fn0.c,
1168 testsuite/libffi.call/cls_9byte2.c,
1169 testsuite/libffi.call/closure_fn2.c,
1170 testsuite/libffi.call/closure_fn4.c,
1171 testsuite/libffi.call/cls_ushort.c,
1172 testsuite/libffi.call/closure_fn6.c,
1173 testsuite/libffi.call/cls_5byte.c,
1174 testsuite/libffi.call/cls_align_pointer.c,
1175 testsuite/libffi.call/cls_7byte.c,
1176 testsuite/libffi.call/cls_align_sint32.c,
1177 testsuite/libffi.special/unwindtest_ffi_call.cc,
1178 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1180 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1182 * aclocal.m4: Regenerated.
1184 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1186 * configure: Regenerate.
1188 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1190 * Makefile.in: Regenerate.
1191 * configure: Regenerate.
1192 * aclocal.m4: Regenerate.
1193 * include/Makefile.in: Regenerate.
1194 * testsuite/Makefile.in: Regenerate.
1196 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1198 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1199 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1200 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1201 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1202 (FFI_CLOSURES): Enable closure support.
1204 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1206 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1207 * configure: Regenerate.
1208 * fficonfig.h.in: Regenerate.
1209 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1210 CFI_OFFSET,CFI_DEF_CFA): New macros.
1211 (ffi_call_SYSV): Add callframe annotation.
1213 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1215 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1216 numerous test suite failures.
1217 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1219 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1221 * Makefile.am (EXTRA_DIST): Bring up to date.
1222 * Makefile.in: Regenerate.
1223 * src/frv/eabi.S: Remove RCS keyword.
1225 2007-04-06 Richard Henderson <rth@redhat.com>
1227 * configure.ac: Tidy target case.
1228 (HAVE_LONG_DOUBLE): Allow the target to override.
1229 * configure: Regenerate.
1230 * include/ffi.h.in: Don't define ffi_type_foo if
1231 LIBFFI_HIDE_BASIC_TYPES is defined.
1232 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1234 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1235 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1236 (ffi_type_longdouble): Special case for Alpha. Don't define
1237 if long double == double.
1239 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1240 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1241 (ffi_call, ffi_closure_osf_inner): Likewise.
1242 (ffi_closure_osf_inner): Likewise. Mark hidden.
1243 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1244 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1245 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1246 (load_table): Handle 128-bit long double.
1248 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1250 2007-04-06 Tom Tromey <tromey@redhat.com>
1253 * README: Fixed bug in example.
1255 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1257 * src/closures.c: Include sys/statfs.h.
1258 (_GNU_SOURCE): Define on Linux.
1259 (FFI_MMAP_EXEC_SELINUX): Define.
1260 (selinux_enabled): New variable.
1261 (selinux_enabled_check): New function.
1262 (is_selinux_enabled): Define.
1265 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1267 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1268 Use 'volatile float sum' to create sum of floats to avoid false
1269 negative due to excess precision on ix86 targets.
1272 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1274 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1276 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1278 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1280 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1281 (ffi_prep_closure_loc): New.
1282 (ffi_prep_raw_closure_loc): New.
1283 (ffi_prep_java_raw_closure_loc): New.
1284 * src/closures.c: New file.
1285 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1286 Replace sflags with exec_offset.
1287 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1288 sub_segment_exec_offset): New macros.
1289 (get_segment_flags, set_segment_flags, check_segment_merge): New
1291 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1292 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1293 destroy_mspace): Use new macros.
1294 (sys_alloc): Silence warning.
1295 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1296 * Makefile.in: Rebuilt.
1297 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1298 terms of ffi_prep_closure_loc.
1299 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1301 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1303 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1305 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1306 the renamed version.
1307 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1308 (ffi_prep_closure): ... this.
1309 * src/pa/ffi.c: Likewise.
1310 * src/cris/ffi.c: Likewise. Adjust.
1311 * src/frv/ffi.c: Likewise.
1312 * src/ia64/ffi.c: Likewise.
1313 * src/mips/ffi.c: Likewise.
1314 * src/powerpc/ffi_darwin.c: Likewise.
1315 * src/s390/ffi.c: Likewise.
1316 * src/sh/ffi.c: Likewise.
1317 * src/sh64/ffi.c: Likewise.
1318 * src/sparc/ffi.c: Likewise.
1319 * src/x86/ffi64.c: Likewise.
1320 * src/x86/ffi.c: Likewise.
1321 (FFI_INIT_TRAMPOLINE): Adjust.
1322 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1323 (ffi_prep_raw_closure): ... this.
1324 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1325 (ffi_prep_closure): ... this.
1326 (flush_icache): Adjust.
1328 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1330 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1333 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1335 * Makefile.am: Add dummy install-pdf target.
1336 * Makefile.in: Regenerate
1338 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1340 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1341 ffi_closure_helper_SYSV): Add long double handling.
1343 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1345 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1346 immediately after bctrl instruction.
1348 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1350 * Makefile.am (all-recursive, install-recursive,
1351 mostlyclean-recursive, clean-recursive, distclean-recursive,
1352 maintainer-clean-recursive): Add missing targets.
1353 * Makefile.in: Rebuilt.
1355 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1357 * configure.ac: Add TARGET for x86_64-*-darwin*.
1358 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1360 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1361 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1362 * configure: Regenerate.
1363 * Makefile.in: Regenerate.
1364 * include/Makefile.in: Regenerate.
1365 * testsuite/Makefile.in: Regenerate.
1366 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1369 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1371 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1374 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1376 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1377 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1379 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1380 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1382 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1383 Sandro Tolaini <tolaini@libero.it>
1385 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1387 * configure: Regenerated.
1388 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1389 (EXTRA_DIST): Add src/x86/darwin.S.
1390 * Makefile.in: Regenerated.
1391 * include/Makefile.in: Regenerated.
1392 * testsuite/Makefile.in: Regenerated.
1394 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1395 X86_WIN32, and additionally align stack to 16 bytes.
1396 * src/x86/darwin.S: New, based on sysv.S.
1397 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1399 2006-09-12 David Daney <ddaney@avtrex.com>
1402 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1403 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1404 * aclocal.m4: Regenerated for automake 1.9.6.
1405 * Makefile.in: Regenerated.
1406 * include/Makefile.in: Regenerated.
1407 * testsuite/Makefile.in: Regenerated.
1409 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1411 * include/ffi_common.h (struct): Revert accidental commit.
1413 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1415 * include/ffi_common.h: Remove lint directives.
1416 * include/ffi.h.in: Likewise.
1418 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1420 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1421 for 32-bit architectures.
1422 * testsuite/libffi.call/return_ul.c: New test case.
1424 2006-07-19 David Daney <ddaney@avtrex.com>
1426 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1427 xfail remains for mips64.
1429 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1431 * Makefile.am: Add install-html target. Add install-html to .PHONY
1432 * Makefile.in: Regenerate.
1433 * aclocal.m4: Regenerate.
1434 * include/Makefile.in: Regenerate.
1435 * testsuite/Makefile.in: Regenerate.
1437 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1439 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1442 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1444 * README: Remove notice about 'Crazy Comments'.
1445 * src/debug.c: Remove lint directives. Cleanup white spaces.
1446 * src/java_raw_api.c: Likewise.
1447 * src/prep_cif.c: Likewise.
1448 * src/raw_api.c: Likewise.
1449 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1451 * src/arm/ffi.c: Remove lint directives.
1452 * src/m32r/ffi.c: Likewise.
1453 * src/pa/ffi.c: Likewise.
1454 * src/powerpc/ffi.c: Likewise.
1455 * src/powerpc/ffi_darwin.c: Likewise.
1456 * src/sh/ffi.c: Likewise.
1457 * src/sh64/ffi.c: Likewise.
1458 * src/x86/ffi.c: Likewise.
1459 * testsuite/libffi.call/float2.c: Likewise.
1460 * testsuite/libffi.call/promotion.c: Likewise.
1461 * testsuite/libffi.call/struct1.c: Likewise.
1463 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1465 * src/pa/hpux32.S: Correct unwind offset calculation for
1467 * src/pa/linux.S: Likewise.
1469 2006-04-12 James E Wilson <wilson@specifix.com>
1472 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1473 (hfa_type_load): Call stf_spill.
1474 (hfa_type_store): Call ldf_fill.
1475 (ffi_call): Adjust calls to above routines. Add local temps for
1478 2006-04-10 Matthias Klose <doko@debian.org>
1480 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1481 directory names containing underscores.
1483 2006-04-07 James E Wilson <wilson@specifix.com>
1485 * testsuite/libffi.call/float4.c: New testcase.
1487 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1488 Andreas Tobler <a.tobler@schweiz.ch>
1490 * Makefile.am: Add PA_HPUX port.
1491 * Makefile.in: Regenerate.
1492 * include/Makefile.in: Likewise.
1493 * testsuite/Makefile.in: Likewise.
1494 * configure.ac: Add PA_HPUX rules.
1495 * configure: Regenerate.
1496 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1497 Add PA_HPUX and PA64_HPUX.
1498 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1499 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1500 (FFI_TYPE_SMALL_STRUCT2): Define.
1501 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1502 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1503 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1504 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1505 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1506 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1507 * src/pa/ffi.c (ROUND_DOWN): Delete.
1508 (fldw, fstw, fldd, fstd): Use '__asm__'.
1509 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1510 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1511 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1512 Simplify incrementing of stack slot variable. Change type of local
1513 'n' to unsigned int.
1514 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1516 (ffi_prep_cif_machdep): Likewise.
1517 (ffi_call): Likewise.
1518 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1519 return type to ffi_status. Simplify incrementing of stack slot
1520 variable. Only copy floating point argument registers when PA_LINUX
1521 is true. Reformat debug statement.
1522 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1523 FFI_TYPE_SMALL_STRUCT8.
1524 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1526 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1527 Add nops to cache flush. Add trampoline for PA_HPUX.
1528 * src/pa/hpux32.S: New file.
1529 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1530 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1531 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1532 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1533 argument type checks so that common argument types appear first.
1534 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1535 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1537 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1539 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1540 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1541 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1542 FFI_TYPE_LONGDOUBLE.
1543 (ffi_prep_args64): Assert using IBM extended double.
1544 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1545 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1546 (ffi_call): Handle FFI_LINUX.
1547 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1548 gpr3 return pointer as for struct return. Handle FFI_LINUX
1549 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1551 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1552 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1553 Don't use r6 as pointer to results, instead use sp offset. Don't
1554 make a special call to load lr with case table address, instead
1555 use offset from previous call.
1556 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1557 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1560 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1562 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1563 passed with FP registers correctly.
1564 (ffi_closure_helper_SYSV): Likewise.
1565 * src/sh64/sysv.S: Likewise.
1567 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1569 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1570 args and userdata unused.
1571 (closure_test_fn1): Mark cif and userdata unused.
1572 (main): Remove unused res.
1574 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1576 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1577 -O2, -O3, -Os and the warning flags -W -Wall.
1578 * testsuite/libffi.special/special.exp: Likewise.
1579 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1580 unused parameter unused for gcc or else do nothing.
1581 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1582 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1583 and userdata unused.
1584 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1585 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1586 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1587 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1588 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1589 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1590 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1591 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1592 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1593 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1594 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1595 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1596 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1597 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1598 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1599 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1600 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1601 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1602 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1603 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1604 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1605 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1606 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1607 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1609 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1611 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1613 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1614 void* to avoid compiler warning.
1616 (cls_struct_align_gn): Mark cif and userdata unused.
1617 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1619 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1621 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1623 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1625 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1627 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1628 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1629 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1631 (main): Cast res_call to silence gcc.
1632 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1634 (main): Cast res_call to silence gcc.
1635 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1637 (main): Cast res_call to silence gcc.
1638 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1640 (main): Cast res_call to silence gcc.
1641 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1643 (main): Cast res_call to silence gcc.
1644 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1646 (main): Cast res_call to silence gcc.
1647 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1649 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1650 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1652 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1653 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1655 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1656 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1658 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1659 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1661 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1662 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1663 and userdata unused.
1664 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1666 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1667 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1668 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1669 Cleanup white spaces.
1670 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1671 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1672 cif and userdata unused.
1673 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1675 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1676 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1677 formatters to silence gcc.
1678 (B_gn): Mark cif and userdata unused.
1679 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1681 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1682 (B_gn): Mark cif and userdata unused.
1683 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1685 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1686 (B_gn): Mark cif and userdata unused.
1687 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1689 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1690 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1691 * testsuite/libffi.call/problem1.c (stub): Likewise.
1692 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1694 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1695 in the last commit for this test case in the test case itself.
1696 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1698 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1699 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1700 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1701 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1702 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1703 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1705 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1707 * src/sh/sysv.S: Fix register numbers in the FDE for
1710 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1712 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1713 declaration to avoid a false negative on ix86. See PR323.
1715 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1717 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1718 and cast integer to void * if needed. Update the pointer to
1719 the FP register saved area correctly.
1721 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1723 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1725 * testsuite/libffi.call/nested_struct4.c: Likewise.
1727 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1729 * testsuite/libffi.call/return_dbl.c: New test case.
1730 * testsuite/libffi.call/return_dbl1.c: Likewise.
1731 * testsuite/libffi.call/return_dbl2.c: Likewise.
1732 * testsuite/libffi.call/return_fl.c: Likewise.
1733 * testsuite/libffi.call/return_fl1.c: Likewise.
1734 * testsuite/libffi.call/return_fl2.c: Likewise.
1735 * testsuite/libffi.call/return_fl3.c: Likewise.
1736 * testsuite/libffi.call/closure_fn6.c: Likewise.
1738 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1740 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1741 here to be used by other test cases too.
1743 * testsuite/libffi.call/nested_struct10.c: New test case.
1744 * testsuite/libffi.call/nested_struct9.c: Likewise.
1745 * testsuite/libffi.call/nested_struct8.c: Likewise.
1746 * testsuite/libffi.call/nested_struct7.c: Likewise.
1747 * testsuite/libffi.call/nested_struct6.c: Likewise.
1748 * testsuite/libffi.call/nested_struct5.c: Likewise.
1749 * testsuite/libffi.call/nested_struct4.c: Likewise.
1751 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1753 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1754 * configure: Rebuilt.
1756 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1758 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1759 instead do the shifting inline.
1760 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1761 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1762 and 8 byte structs, for the remaining struct sizes don't call
1763 __lshrdi3, instead do the shifting inline.
1765 2005-12-07 Thiemo Seufer <ths@networkno.de>
1767 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1768 missing parentheses.
1769 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1770 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1771 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1772 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1773 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1775 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1777 (ffi_prep_closure): Improve trampoline instruction scheduling.
1778 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1780 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1782 * src/powerpc/ffi.c: Formatting.
1783 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1784 (ffi_prep_args64): Likewise.
1786 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1788 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1789 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1791 2005-09-26 Tom Tromey <tromey@redhat.com>
1793 * testsuite/libffi.call/float1.c (value_type): New typedef.
1794 (CANARY): New define.
1795 (main): Check for result buffer overflow.
1796 * src/powerpc/linux64.S: Handle linux64 long double returns.
1797 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1798 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1800 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1803 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1805 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1807 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1809 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1810 (AH_BOTTOM): Add FFI_HIDDEN definition.
1811 * configure: Rebuilt.
1812 * fficonfig.h.in: Rebuilt.
1813 * src/powerpc/ffi.c (hidden): Remove.
1814 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1815 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1816 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1817 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1818 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1819 add FFI_HIDDEN to its prototype.
1820 (ffi_closure_SYSV_inner): New.
1821 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1822 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1824 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1827 * configure: Rebuilt.
1828 * configure.ac: Handle i*86-*-gnu*.
1830 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1832 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1833 DW_CFA_offset_extended_sf rather than
1834 DW_CFA_GNU_negative_offset_extended.
1835 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1837 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1839 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1841 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1842 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1843 partially on register.
1844 (ffi_closure_helper_SYSV): Likewise.
1845 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1847 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1849 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1851 * src/sh64/ffi.c (simple_type): Remove.
1852 (return_type): Handle small structures correctly.
1853 (ffi_prep_args): Likewise.
1854 (ffi_call): Likewise.
1855 (ffi_closure_helper_SYSV): Likewise.
1856 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1857 Emit position independent code if PIC and remove wrong datalabel
1858 prefixes from EH data.
1860 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1862 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1863 * Makefile.in: Regenerate.
1864 * include/Makefile.in: Likewise.
1865 * testsuite/Makefile.in: Likewise.
1866 * configure.ac: Add POWERPC_FREEBSD rules.
1867 * configure: Regenerate.
1868 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1869 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1870 * src/powerpc/ffi.c: Add flags to handle small structure returns
1872 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1874 (ffi_closure_helper_SYSV): Likewise.
1875 * src/powerpc/ppc_closure.S: Add return types for small structures.
1876 * src/powerpc/sysv.S: Add bits to handle small structures for
1879 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1881 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1882 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1883 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1885 2005-07-05 Randolph Chung <tausq@debian.org>
1887 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1888 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1889 structures. Kill compilation warnings.
1890 (ffi_closure_inner_LINUX): Print return values as hex in debug
1891 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1892 Properly handle 5-7 byte structure returns.
1893 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1894 (FFI_TYPE_SMALL_STRUCT2): Remove.
1895 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1896 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1897 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1898 (checksmst1, checksmst2): Remove.
1899 (checksmst3): Optimize handling of 3-byte struct returns.
1900 (checksmst567): Properly handle 5-7 byte struct returns.
1902 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1905 * src/mips/n32.S: Enforce PIC code.
1906 * src/mips/o32.S: Likewise.
1908 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1910 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1911 * configure: Regenerate.
1913 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1915 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1916 to call ffi_closure_helper_SYSV. Append @local instead.
1917 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1919 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1921 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1922 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1923 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1924 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1925 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1927 2005-05-09 Mike Stump <mrs@apple.com>
1929 * configure: Regenerate.
1931 2005-05-08 Richard Henderson <rth@redhat.com>
1934 * src/alpha/osf.S: Update unwind into to match code.
1936 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1937 Richard Henderson <rth@redhat.com>
1939 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1941 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1942 (ffi_prep_closure): Set carry bit if sse-used flag set.
1943 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1944 Only load sse registers if ssecount non-zero.
1945 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1947 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1949 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1950 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1951 * configure: Regenerate.
1953 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1955 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1956 have Tcl8.3-compatible intermediate variable.
1958 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1959 Hans-Peter Nilsson <hp@axis.com>
1961 * Makefile.am: Add CRIS support.
1962 * configure.ac: Likewise.
1963 * Makefile.in, configure, testsuite/Makefile.in,
1964 include/Makefile.in: Regenerate.
1965 * src/cris: New directory.
1966 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1967 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1969 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1970 \r?\n in output tests.
1972 2005-04-12 Mike Stump <mrs@apple.com>
1974 * configure: Regenerate.
1976 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1978 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1980 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1982 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1984 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1985 to 64 bits in ILP32 mode.
1986 Load 64 bits even for short data.
1988 2005-03-23 Mike Stump <mrs@apple.com>
1990 * src/powerpc/darwin.S: Update for -m64 multilib.
1991 * src/powerpc/darwin_closure.S: Likewise.
1993 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1995 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1996 Do not set tool_include_dir.
1997 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1999 * include/Makefile.am: Set gcc_version and toollibffidir.
2000 * include/Makefile.in: Regenerate.
2002 2005-02-22 Andrew Haley <aph@redhat.com>
2004 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2005 odd-numbered register pairs for 64-bit integer types.
2007 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2010 * testsuite/libffi.call/return_ll1.c: New test case.
2012 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2014 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2015 * testsuite/libffi.call/float.c: Ditto.
2016 * testsuite/libffi.call/float2.c: Ditto.
2017 * testsuite/libffi.call/float3.c: Ditto.
2019 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2021 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2023 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2025 * testsuite/libffi.special/special.exp (cxx_options): Add
2028 2004-12-31 Richard Henderson <rth@redhat.com>
2030 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2031 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2032 offset parameters with a type parameter; deduce size and structure
2033 alignment. Update all users.
2035 2004-12-31 Richard Henderson <rth@redhat.com>
2037 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2038 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2039 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2040 into ffi_prep_closure.
2041 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2044 2004-12-27 Richard Henderson <rth@redhat.com>
2046 * src/x86/unix64.S: Fix typo in unwind info.
2048 2004-12-25 Richard Henderson <rth@redhat.com>
2050 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2051 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2052 (merge_classes): Check for it.
2054 (classify_argument): Pass byte_offset by value; perform all updates
2056 (examine_argument): Add classes argument; handle
2057 X86_64_COMPLEX_X87_CLASS.
2058 (ffi_prep_args): Merge into ...
2059 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2060 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2061 (ffi_fill_return_value): Remove.
2062 (ffi_prep_closure): Remove dead assert.
2063 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2064 Rewrite to use struct register_args instead of va_list. Create
2065 flags for handling structure returns.
2066 * src/x86/unix64.S: Remove dead strings.
2067 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2068 stack frame with ffi_call. Handle structure returns properly.
2069 (float2sse, floatfloat2sse, double2sse): Remove.
2070 (sse2float, sse2double, sse2floatfloat): Remove.
2071 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2072 to handle structure returns properly.
2074 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2076 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2078 * Makefile.in: Regenerated.
2080 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2082 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2083 * configure, aclocal.m4, Makefile.in: Regenerate.
2084 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2086 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2088 * configure: Regenerate for libtool change.
2090 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2092 * configure: Regenerate for libtool reversion.
2094 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2096 * configure: Regenerate for libtool change.
2098 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2100 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2102 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2104 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2105 of jal. Use an absolute encoding for the frame information.
2107 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2109 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2110 * acinclude.m4: Delete logic for sincludes.
2111 * aclocal.m4, Makefile.in, configure: Regenerate.
2112 * include/Makefile: Likewise.
2113 * testsuite/Makefile: Likewise.
2115 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2117 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2118 on a 8-byte boundary.
2119 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2121 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2123 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2124 long long values. Round stack allocation to a multiple of 8 bytes
2125 for ATPCS compatibility.
2126 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2127 names. Handle returning long long types. Add Thumb and interworking
2128 support. Improve soft-float code.
2130 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2132 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2133 (libffi_exit): New function.
2134 (libffi_init): Build the testglue wrapper if needed.
2136 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2139 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2141 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2143 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2145 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2147 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2148 * testsuite/libffi.call/float3.c: New test case.
2150 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2152 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2153 the function returning a structure pointed with R2.
2154 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2155 the structure return value if T bit set. Emit position
2156 independent code and EH data if PIC.
2158 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2160 * Makefile.am: Add m32r support.
2161 * configure.ac: Likewise.
2162 * Makefile.in: Regenerate.
2163 * confiugre: Regenerate.
2164 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2165 (uint64, sint64, double, longdouble)
2166 * src/m32r: New directory.
2167 * src/m32r/ffi.c: New file.
2168 * src/m32r/sysv.S: Likewise.
2169 * src/m32r/ffitarget.h: Likewise.
2171 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2173 * testsuite/libffi.call/negint.c: New test case.
2175 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2178 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2179 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2180 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2183 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2185 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2186 * testsuite/libffi.call/promotion.c: Likewise.
2187 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2189 * testsuite/libffi.call/return_sc.c: Likewise.
2190 * testsuite/libffi.call/return_uc.c: Likewise.
2192 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2194 * src/powerpc/darwin.S: Fix comments and identation.
2195 * src/powerpc/darwin_closure.S: Likewise.
2197 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2199 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2200 (ffi_prep_args): Handle longdouble arguments.
2201 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2203 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2204 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2206 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2207 * src/types.c: Defined longdouble size and alignment for darwin.
2209 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2211 * src/powerpc/aix.S: Remove whitespaces.
2212 * src/powerpc/aix_closure.S: Likewise.
2213 * src/powerpc/asm.h: Likewise.
2214 * src/powerpc/ffi.c: Likewise.
2215 * src/powerpc/ffitarget.h: Likewise.
2216 * src/powerpc/linux64.S: Likewise.
2217 * src/powerpc/linux64_closure.S: Likewise.
2218 * src/powerpc/ppc_closure.S: Likewise.
2219 * src/powerpc/sysv.S: Likewise.
2221 2004-08-30 Anthony Green <green@redhat.com>
2223 * Makefile.am: Add frv support.
2224 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2225 * configure.ac: Read configure.host.
2226 * configure.in: Read configure.host.
2227 * configure.host: New file. frv-elf needs libgloss.
2228 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2229 alignment. This is needed to frv and shouldn't harm the others.
2230 * include/ffi_common.h (ALIGN_DOWN): New macro.
2231 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2233 2004-08-24 David Daney <daney@avtrex.com>
2235 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2236 * testsuite/libffi.call/closure_fn1.c: Likewise.
2237 * testsuite/libffi.call/closure_fn2.c Likewise.
2238 * testsuite/libffi.call/closure_fn3.c: Likewise.
2239 * testsuite/libffi.call/closure_fn4.c: Likewise.
2240 * testsuite/libffi.call/closure_fn5.c: Likewise.
2241 * testsuite/libffi.call/cls_18byte.c: Likewise.
2242 * testsuite/libffi.call/cls_19byte.c: Likewise.
2243 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2244 * testsuite/libffi.call/cls_20byte.c: Likewise.
2245 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2246 * testsuite/libffi.call/cls_24byte.c: Likewise.
2247 * testsuite/libffi.call/cls_2byte.c: Likewise.
2248 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2249 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2250 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2251 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2252 * testsuite/libffi.call/cls_4byte.c: Likewise.
2253 * testsuite/libffi.call/cls_64byte.c: Likewise.
2254 * testsuite/libffi.call/cls_6byte.c: Likewise.
2255 * testsuite/libffi.call/cls_7byte.c: Likewise.
2256 * testsuite/libffi.call/cls_8byte.c: Likewise.
2257 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2258 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2259 * testsuite/libffi.call/cls_align_double.c: Likewise.
2260 * testsuite/libffi.call/cls_align_float.c: Likewise.
2261 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2262 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2263 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2264 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2265 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2266 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2267 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2268 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2269 * testsuite/libffi.call/cls_double.c: Likewise.
2270 * testsuite/libffi.call/cls_float.c: Likewise.
2271 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2272 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2273 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2274 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2275 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2276 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2277 * testsuite/libffi.call/cls_schar.c: Likewise.
2278 * testsuite/libffi.call/cls_sint.c: Likewise.
2279 * testsuite/libffi.call/cls_sshort.c: Likewise.
2280 * testsuite/libffi.call/cls_uchar.c: Likewise.
2281 * testsuite/libffi.call/cls_uint.c: Likewise.
2282 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2283 * testsuite/libffi.call/cls_ushort.c: Likewise.
2284 * testsuite/libffi.call/nested_struct.c: Likewise.
2285 * testsuite/libffi.call/nested_struct1.c: Likewise.
2286 * testsuite/libffi.call/nested_struct2.c: Likewise.
2287 * testsuite/libffi.call/nested_struct3.c: Likewise.
2288 * testsuite/libffi.call/problem1.c: Likewise.
2289 * testsuite/libffi.special/unwindtest.cc: Likewise.
2290 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2292 * testsuite/libffi.call/cls_16byte.c: Likewise.
2293 * testsuite/libffi.call/cls_5byte.c: Likewise.
2295 2004-08-23 David Daney <daney@avtrex.com>
2298 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2299 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2300 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2301 parameters and return types.
2302 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2303 (ffi_prep_closure): Ditto.
2304 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2305 alignment calculations.
2306 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2307 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2309 2004-08-14 Casey Marshall <csm@gnu.org>
2311 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2312 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2313 integer (O32 ABI only).
2314 (ffi_prep_closure): new function.
2315 (ffi_closure_mips_inner_O32): new function.
2316 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2317 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2318 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2319 64 bit integers correctly.
2320 (ffi_closure_O32): new function.
2321 Added DWARF-2 unwind info for both functions.
2323 2004-08-10 Andrew Haley <aph@redhat.com>
2325 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2327 2004-08-01 Robert Millan <robertmh@gnu.org>
2329 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2330 * configure: Regenerate.
2332 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2334 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2335 and mmap() explicitly instead of relying on preset autoconf cache
2337 * aclocal.m4: Regenerate.
2338 * configure: Regenerate.
2340 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2342 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2343 (ffi_check_float_struct): Remove unused prototype.
2345 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2347 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2348 character on Darwin, use '\n\t' instead.
2350 2004-06-26 Matthias Klose <doko@debian.org>
2352 * libtool-version: Fix typo in revision/age.
2354 2004-06-17 Matthias Klose <doko@debian.org>
2356 * libtool-version: New.
2357 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2358 * Makefile.in: Regenerate.
2360 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2362 * Makefile.am: Remove useless multilib rules.
2363 * Makefile.in: Regenerate.
2364 * aclocal.m4: Regenerate with automake 1.8.5.
2365 * configure.ac: Remove useless multilib configury.
2366 * configure: Regenerate.
2368 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2370 * .cvsignore: New file.
2372 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2374 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2376 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2379 2004-06-06 Sean McNeil <sean@mcneil.com>
2381 * configure.ac: Add x86_64-*-freebsd* support.
2382 * configure: Regenerate.
2384 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2387 * configure.ac: Test for existence of mmap and sys/mman.h before
2388 checking blacklist. Fix suggested by Jim Wilson.
2389 * configure: Regenerate.
2391 2004-04-26 Matt Austern <austern@apple.com>
2393 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2395 * src/powerpc/darwin_closure.S: Likewise.
2397 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2399 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2400 error. Reported by Thomas Heller <theller@python.net>.
2401 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2402 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2404 2004-03-20 Matthias Klose <doko@debian.org>
2406 * src/pa/linux.S: Fix typo.
2408 2004-03-19 Matthias Klose <doko@debian.org>
2410 * Makefile.am: Update.
2411 * Makefile.in: Regenerate.
2412 * src/pa/ffi.h.in: Remove.
2413 * src/pa/ffitarget.h: New file.
2415 2004-02-10 Randolph Chung <tausq@debian.org>
2417 * Makefile.am: Add PA support.
2418 * Makefile.in: Regenerate.
2419 * include/Makefile.in: Regenerate.
2420 * configure.ac: Add PA target.
2421 * configure: Regenerate.
2422 * src/pa/ffi.c: New file.
2423 * src/pa/ffi.h.in: Add PA support.
2424 * src/pa/linux.S: New file.
2425 * prep_cif.c: Add PA support.
2427 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2429 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2431 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2432 with ecif->cif->flags.
2433 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2435 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2436 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2437 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2438 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2440 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2442 * configure.in: Rename file to ...
2443 * configure.ac: ... this.
2444 * fficonfig.h.in: Regenerate.
2445 * Makefile.in: Regenerate.
2446 * include/Makefile.in: Regenerate.
2447 * testsuite/Makefile.in: Regenerate.
2449 2004-03-12 Matt Austern <austern@apple.com>
2451 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2452 changes in EH format resulting from addition of linkonce support.
2453 * src/powerpc/darwin_closure.S: Likewise.
2455 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2456 Paolo Bonzini <bonzini@gnu.org>
2458 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2459 Remove VPATH. Remove rules for object files. Remove multilib support.
2460 (AM_CCASFLAGS): Add.
2461 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2462 (AC_PREREQ): Bump version to 2.59.
2463 (AC_INIT): Fill with version info and bug address.
2464 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2465 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2466 De-precious CC so that the right flags are passed down to multilibs.
2467 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2468 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2469 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2470 * configure: Rebuilt.
2471 * aclocal.m4: Likewise.
2472 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2473 * fficonfig.h.in: Likewise.
2475 2004-03-11 Andreas Schwab <schwab@suse.de>
2477 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2478 arguments from fp registers only for the first 8 parameter slots.
2479 Don't convert a float parameter when passed in memory.
2481 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2483 * configure: Regenerate for config/accross.m4 correction.
2485 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2487 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2488 ecif->cif->bytes to bytes.
2489 (ffi_prep_cif_machdep): Add braces around nested if statement.
2491 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2493 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2495 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2496 (ffi_closure_helper_LINUX64): Fix typo.
2497 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2499 * testsuite/libffi.call/float.c: Likewise.
2500 * testsuite/libffi.call/float2.c: Likewise.
2502 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2504 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2505 long double function return and long double arg handling.
2506 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2507 Use "end_pfr" instead of "nf". Correct long double handling.
2509 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2511 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2512 space for long double return value. Adjust stack frame and offsets.
2513 Load f2 long double return.
2515 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2517 * src/types.c: Use 16 byte long double for POWERPC64.
2519 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2521 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2522 when the structure return address is passed in %o0.
2523 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2524 (ffi_v9_layout_struct): Align the field following a nested structure
2525 on a word boundary. Use memmove instead of memcpy.
2526 (ffi_call): Update call to ffi_V9_return_struct.
2527 (ffi_prep_closure): Define 'ctx' only for V8.
2528 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2529 and ffi_closure_sparc_inner_v9.
2530 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2531 Always skip the structure return address. For structures and long
2532 doubles, copy the argument directly.
2533 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2534 if required. Shift the maximum floating-point slot accordingly. For
2535 big structures, copy the argument directly; otherwise, left-justify the
2536 argument and call ffi_v9_layout_struct to lay out the structure on
2538 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2539 (ffi_closure_v8): Pass the structure return address. Update call to
2540 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2541 Skip the 'unimp' insn when returning long doubles and structures.
2542 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2543 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2544 FFI_TYPE_INT handling. Load structures both in integers and
2545 floating-point registers on return.
2546 * README: Update status of the SPARC port.
2548 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2550 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2552 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2554 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2556 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2557 value as of type ffi_arg, not unsigned int.
2559 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2561 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2564 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2566 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2569 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2571 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2574 2003-12-10 Richard Henderson <rth@redhat.com>
2576 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2577 size_t instead of int.
2579 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2581 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2582 * testsuite/libffi.call/many_win32.c (main): Replace variable
2583 int i with unsigned long ul.
2585 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2586 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2587 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2588 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2589 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2590 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2591 * testsuite/libffi.call/cls_align_float.c: Likewise.
2592 * testsuite/libffi.call/cls_align_double.c: Likewise.
2593 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2594 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2596 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2599 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2600 Align arguments to 32 bits.
2602 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2605 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2606 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2607 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2608 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2609 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2610 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2612 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2614 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2616 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2617 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2619 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2621 * Makefile.in: Rebuilt.
2622 * configure: Likewise.
2623 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2626 2003-11-21 Andreas Jaeger <aj@suse.de>
2627 Andreas Tobler <a.tobler@schweiz.ch>
2629 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2630 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2631 * Makefile.in: Rebuilt.
2632 * aclocal.m4: Likewise.
2633 * configure: Likewise.
2634 * fficonfig.h.in: Likewise.
2635 * testsuite/lib/libffi-dg.exp: Add include dir.
2636 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2637 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2638 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2639 for ffi_closure if available.
2640 * testsuite/libffi.call/closure_fn1.c: Likewise.
2641 * testsuite/libffi.call/closure_fn2.c: Likewise.
2642 * testsuite/libffi.call/closure_fn3.c: Likewise.
2643 * testsuite/libffi.call/closure_fn4.c: Likewise.
2644 * testsuite/libffi.call/closure_fn5.c: Likewise.
2645 * testsuite/libffi.call/cls_12byte.c: Likewise.
2646 * testsuite/libffi.call/cls_16byte.c: Likewise.
2647 * testsuite/libffi.call/cls_18byte.c: Likewise.
2648 * testsuite/libffi.call/cls_19byte.c: Likewise.
2649 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2650 * testsuite/libffi.call/cls_20byte.c: Likewise.
2651 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2652 * testsuite/libffi.call/cls_24byte.c: Likewise.
2653 * testsuite/libffi.call/cls_2byte.c: Likewise.
2654 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2655 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2656 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2657 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2658 * testsuite/libffi.call/cls_4byte.c: Likewise.
2659 * testsuite/libffi.call/cls_5byte.c: Likewise.
2660 * testsuite/libffi.call/cls_64byte.c: Likewise.
2661 * testsuite/libffi.call/cls_6byte.c: Likewise.
2662 * testsuite/libffi.call/cls_7byte.c: Likewise.
2663 * testsuite/libffi.call/cls_8byte.c: Likewise.
2664 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2665 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2666 * testsuite/libffi.call/cls_double.c: Likewise.
2667 * testsuite/libffi.call/cls_float.c: Likewise.
2668 * testsuite/libffi.call/cls_schar.c: Likewise.
2669 * testsuite/libffi.call/cls_sint.c: Likewise.
2670 * testsuite/libffi.call/cls_sshort.c: Likewise.
2671 * testsuite/libffi.call/cls_uchar.c: Likewise.
2672 * testsuite/libffi.call/cls_uint.c: Likewise.
2673 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2674 * testsuite/libffi.call/cls_ushort.c: Likewise.
2675 * testsuite/libffi.call/nested_struct.c: Likewise.
2676 * testsuite/libffi.call/nested_struct1.c: Likewise.
2677 * testsuite/libffi.call/nested_struct2.c: Likewise.
2678 * testsuite/libffi.call/nested_struct3.c: Likewise.
2679 * testsuite/libffi.call/problem1.c: Likewise.
2680 * testsuite/libffi.special/unwindtest.cc: Likewise.
2682 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2684 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2686 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2688 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2689 Add -lgcc_s to additional flags.
2691 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2693 * configure.in, include/Makefile.am: PR libgcj/11147, install
2694 the ffitarget.h header file in a gcc versioned and target
2696 * configure: Regenerated.
2697 * Makefile.in, include/Makefile.in: Likewise.
2698 * testsuite/Makefile.in: Likewise.
2700 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2702 * testsuite/libffi.call/closure_fn0.c: Print result and check
2703 with dg-output to make debugging easier.
2704 * testsuite/libffi.call/closure_fn1.c: Likewise.
2705 * testsuite/libffi.call/closure_fn2.c: Likewise.
2706 * testsuite/libffi.call/closure_fn3.c: Likewise.
2707 * testsuite/libffi.call/closure_fn4.c: Likewise.
2708 * testsuite/libffi.call/closure_fn5.c: Likewise.
2709 * testsuite/libffi.call/cls_12byte.c: Likewise.
2710 * testsuite/libffi.call/cls_16byte.c: Likewise.
2711 * testsuite/libffi.call/cls_18byte.c: Likewise.
2712 * testsuite/libffi.call/cls_19byte.c: Likewise.
2713 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2714 * testsuite/libffi.call/cls_20byte.c: Likewise.
2715 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2716 * testsuite/libffi.call/cls_24byte.c: Likewise.
2717 * testsuite/libffi.call/cls_2byte.c: Likewise.
2718 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2719 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2720 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2721 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2722 * testsuite/libffi.call/cls_4byte.c: Likewise.
2723 * testsuite/libffi.call/cls_5byte.c: Likewise.
2724 * testsuite/libffi.call/cls_64byte.c: Likewise.
2725 * testsuite/libffi.call/cls_6byte.c: Likewise.
2726 * testsuite/libffi.call/cls_7byte.c: Likewise.
2727 * testsuite/libffi.call/cls_8byte.c: Likewise.
2728 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2729 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2730 * testsuite/libffi.call/cls_double.c: Likewise.
2731 * testsuite/libffi.call/cls_float.c: Likewise.
2732 * testsuite/libffi.call/cls_schar.c: Likewise.
2733 * testsuite/libffi.call/cls_sint.c: Likewise.
2734 * testsuite/libffi.call/cls_sshort.c: Likewise.
2735 * testsuite/libffi.call/cls_uchar.c: Likewise.
2736 * testsuite/libffi.call/cls_uint.c: Likewise.
2737 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2738 * testsuite/libffi.call/cls_ushort.c: Likewise.
2739 * testsuite/libffi.call/problem1.c: Likewise.
2741 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2744 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2746 * testsuite/libffi.call/cls_9byte2.c: New test case.
2747 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2748 * testsuite/libffi.call/cls_64byte.c: Likewise.
2749 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2750 * testsuite/libffi.call/cls_19byte.c: Likewise.
2751 * testsuite/libffi.call/cls_18byte.c: Likewise.
2752 * testsuite/libffi.call/closure_fn4.c: Likewise.
2753 * testsuite/libffi.call/closure_fn5.c: Likewise.
2754 * testsuite/libffi.call/cls_schar.c: Likewise.
2755 * testsuite/libffi.call/cls_sint.c: Likewise.
2756 * testsuite/libffi.call/cls_sshort.c: Likewise.
2757 * testsuite/libffi.call/nested_struct2.c: Likewise.
2758 * testsuite/libffi.call/nested_struct3.c: Likewise.
2760 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2762 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2763 * testsuite/libffi.call/cls_uchar.c: Likewise.
2764 * testsuite/libffi.call/cls_uint.c: Likewise.
2765 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2766 * testsuite/libffi.call/cls_ushort.c: Likewise.
2767 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2769 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2771 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2774 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2776 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2777 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2779 2003-10-22 David Daney <ddaney@avtrex.com>
2781 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2782 __attribute__((__mode__(__SI__))) and friends.
2784 2003-10-22 Andreas Schwab <schwab@suse.de>
2786 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2788 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2790 * configure.in: AC_LINK_FILES(ffitarget.h).
2791 * configure: Regenerate.
2792 * Makefile.in: Likewise.
2793 * include/Makefile.in: Likewise.
2794 * testsuite/Makefile.in: Likewise.
2795 * fficonfig.h.in: Likewise.
2797 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2798 Richard Henderson <rth@redhat.com>
2800 Avoid that ffi.h includes fficonfig.h.
2802 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2803 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2804 (TARGET_SRC_MIPS_SGI): Removed.
2805 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2806 (MIPS_SGI): Removed.
2807 (CLEANFILES): Removed.
2808 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2810 * acconfig.h: Removed.
2811 * configure.in: Compute sizeofs only for double and long double.
2812 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2813 into AC_DEFINE instead of using acconfig.h. Create
2814 include/ffitarget.h instead of include/fficonfig.h. Rename
2815 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2816 AC_DEFINE EH_FRAME_FLAGS.
2817 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2818 (hack_DATA): Add ffitarget.h.
2819 * include/ffi.h.in: Remove all system specific definitions.
2820 Declare raw API even if it is not installed, why bother?
2821 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2822 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2823 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2824 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2825 * include/ffi_common.h (bool): Do not define.
2826 (ffi_assert): Accept failed assertion.
2827 (ffi_type_test): Return void and accept file/line.
2828 (FFI_ASSERT): Pass stringized failed assertion.
2829 (FFI_ASSERT_AT): New macro.
2830 (FFI_ASSERT_VALID_TYPE): New macro.
2831 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2832 UINT64, SINT64): Define here with gcc's __attribute__ macro
2834 (FLOAT32, ALIGN): Define here instead of in ffi.h
2835 * include/ffi-mips.h: Removed. Its content moved to
2836 src/mips/ffitarget.h after separating assembly and C sections.
2837 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2838 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2839 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2840 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2841 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2842 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2843 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2844 (ffi_type_test): Rewritten.
2845 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2846 FFI_ASSERT_VALID_TYPE.
2847 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2848 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2849 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2850 src/s390/ffitarget.h, src/sh/ffitarget.h,
2851 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2852 src/x86/ffitarget.h: New files.
2853 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2854 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2855 src/powerpc/aix.S, src/powerpc/darwin.S,
2856 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2857 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2858 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2859 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2860 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2863 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2865 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2866 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2868 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2870 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2871 Used when FFI_DEBUG = 1.
2873 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2875 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2878 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2880 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2881 FFI_MIPS_O32 for O32 ABI.
2883 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2885 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2886 SPARC64. Cleanup whitespaces.
2888 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2890 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2891 strongarm, xscale. Cleanup whitespaces.
2892 * testsuite/libffi.call/closure_fn1.c: Likewise.
2893 * testsuite/libffi.call/closure_fn2.c: Likewise.
2894 * testsuite/libffi.call/closure_fn3.c: Likewise.
2895 * testsuite/libffi.call/cls_12byte.c: Likewise.
2896 * testsuite/libffi.call/cls_16byte.c: Likewise.
2897 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2898 * testsuite/libffi.call/cls_20byte.c: Likewise.
2899 * testsuite/libffi.call/cls_24byte.c: Likewise.
2900 * testsuite/libffi.call/cls_2byte.c: Likewise.
2901 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2902 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2903 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2904 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2905 * testsuite/libffi.call/cls_4byte.c: Likewise.
2906 * testsuite/libffi.call/cls_5byte.c: Likewise.
2907 * testsuite/libffi.call/cls_6byte.c: Likewise.
2908 * testsuite/libffi.call/cls_7byte.c: Likewise.
2909 * testsuite/libffi.call/cls_8byte.c: Likewise.
2910 * testsuite/libffi.call/cls_double.c: Likewise.
2911 * testsuite/libffi.call/cls_float.c: Likewise.
2912 * testsuite/libffi.call/cls_uchar.c: Likewise.
2913 * testsuite/libffi.call/cls_uint.c: Likewise.
2914 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2915 * testsuite/libffi.call/cls_ushort.c: Likewise.
2916 * testsuite/libffi.call/nested_struct.c: Likewise.
2917 * testsuite/libffi.call/nested_struct1.c: Likewise.
2918 * testsuite/libffi.call/problem1.c: Likewise.
2919 * testsuite/libffi.special/unwindtest.cc: Likewise.
2920 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2922 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2924 * src/powerpc/aix.S: Cleanup whitespaces.
2925 * src/powerpc/aix_closure.S: Likewise.
2927 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2929 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2930 * src/powerpc/darwin_closure.S: Likewise.
2931 * src/powerpc/ffi_darwin.c: Likewise.
2933 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2934 David Edelsohn <edelsohn@gnu.org>
2936 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2937 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2939 * src/powerpc/darwin_closure.S: Likewise.
2940 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2941 according to the Darwin/AIX ABI.
2942 (ffi_prep_cif_machdep): Likewise.
2943 (ffi_closure_helper_DARWIN): Likewise.
2944 Remove the outgoing parameter stack logic. Simplify the evaluation
2945 of the different CASE types.
2946 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2947 statement in the trampoline code.
2949 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2951 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2952 for the register size.
2953 (ffi_closure_helper_SYSV): Handle the structure return value
2955 (ffi_closure_helper_SYSV): Return the appropriate type when
2956 the registers are used for the structure return value.
2957 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2958 the 64-bit return value. Update copyright years.
2960 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2962 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2963 srcdir for ffi_mips.h.
2965 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2967 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2969 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2970 placement of float result.
2971 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2972 cast of "resp" for big-endian 64 bit machines.
2974 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2976 * src/types.c (double, longdouble): Merge identical SH and ARM
2977 typedefs, and add POWERPC64.
2978 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2979 struct split over gpr and rest.
2980 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2981 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2983 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2985 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2988 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2990 * configure: Regenerate.
2992 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2994 * Makefile.am: Remove build rules for ffitest.
2995 * Makefile.in: Rebuilt.
2997 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2999 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3000 about implicit declaration of abort().
3002 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3004 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3005 * Makefile.in: Rebuilt.
3006 * configure.in: Add dejagnu test framework.
3007 * configure: Rebuilt.
3009 * testsuite/Makefile.am: New file.
3010 * testsuite/Makefile.in: Built
3011 * testsuite/lib/libffi-dg.exp: New file.
3012 * testsuite/config/default.exp: Likewise.
3013 * testsuite/libffi.call/call.exp: Likewise.
3014 * testsuite/libffi.call/ffitest.h: Likewise.
3015 * testsuite/libffi.call/closure_fn0.c: Likewise.
3016 * testsuite/libffi.call/closure_fn1.c: Likewise.
3017 * testsuite/libffi.call/closure_fn2.c: Likewise.
3018 * testsuite/libffi.call/closure_fn3.c: Likewise.
3019 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3020 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3021 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3022 * testsuite/libffi.call/cls_2byte.c: Likewise.
3023 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3024 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3025 * testsuite/libffi.call/cls_4byte.c: Likewise.
3026 * testsuite/libffi.call/cls_5byte.c: Likewise.
3027 * testsuite/libffi.call/cls_6byte.c: Likewise.
3028 * testsuite/libffi.call/cls_7byte.c: Likewise.
3029 * testsuite/libffi.call/cls_8byte.c: Likewise.
3030 * testsuite/libffi.call/cls_12byte.c: Likewise.
3031 * testsuite/libffi.call/cls_16byte.c: Likewise.
3032 * testsuite/libffi.call/cls_20byte.c: Likewise.
3033 * testsuite/libffi.call/cls_24byte.c: Likewise.
3034 * testsuite/libffi.call/cls_double.c: Likewise.
3035 * testsuite/libffi.call/cls_float.c: Likewise.
3036 * testsuite/libffi.call/cls_uchar.c: Likewise.
3037 * testsuite/libffi.call/cls_uint.c: Likewise.
3038 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3039 * testsuite/libffi.call/cls_ushort.c: Likewise.
3040 * testsuite/libffi.call/float.c: Likewise.
3041 * testsuite/libffi.call/float1.c: Likewise.
3042 * testsuite/libffi.call/float2.c: Likewise.
3043 * testsuite/libffi.call/many.c: Likewise.
3044 * testsuite/libffi.call/many_win32.c: Likewise.
3045 * testsuite/libffi.call/nested_struct.c: Likewise.
3046 * testsuite/libffi.call/nested_struct1.c: Likewise.
3047 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3048 * testsuite/libffi.call/problem1.c: Likewise.
3049 * testsuite/libffi.call/promotion.c: Likewise.
3050 * testsuite/libffi.call/return_ll.c: Likewise.
3051 * testsuite/libffi.call/return_sc.c: Likewise.
3052 * testsuite/libffi.call/return_uc.c: Likewise.
3053 * testsuite/libffi.call/strlen.c: Likewise.
3054 * testsuite/libffi.call/strlen_win32.c: Likewise.
3055 * testsuite/libffi.call/struct1.c: Likewise.
3056 * testsuite/libffi.call/struct2.c: Likewise.
3057 * testsuite/libffi.call/struct3.c: Likewise.
3058 * testsuite/libffi.call/struct4.c: Likewise.
3059 * testsuite/libffi.call/struct5.c: Likewise.
3060 * testsuite/libffi.call/struct6.c: Likewise.
3061 * testsuite/libffi.call/struct7.c: Likewise.
3062 * testsuite/libffi.call/struct8.c: Likewise.
3063 * testsuite/libffi.call/struct9.c: Likewise.
3064 * testsuite/libffi.special/special.exp: New file.
3065 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3066 * testsuite/libffi.special/unwindtest.cc: Likewise.
3069 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3071 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3074 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3076 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3078 (ffi_closure_helper_LINUX64): Likewise.
3079 * src/powerpc/linux64.S: Remove code writing to parm save area.
3080 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3081 address in lr from ffi_closure_helper_LINUX64 call to calculate
3082 table address. Optimize function tail.
3084 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3086 * src/sparc/ffi.c: Handle all floating point registers.
3087 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3089 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3091 * README: Note that libffi is not part of GCC. Update the project
3094 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3096 * src/powerpc/ppc_closure.S: Include ffi.h.
3098 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3100 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3101 Use C style comments.
3103 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3105 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3106 * Makefile.in: Regenerate.
3107 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3108 * configure: Regenerate.
3109 * include/ffi.h.in: Add SHmedia support.
3110 * src/sh64/ffi.c: New file.
3111 * src/sh64/sysv.S: New file.
3113 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3115 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3116 should be read-only.
3117 * configure: Rebuilt.
3118 * fficonfig.h.in: Rebuilt.
3119 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3120 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3121 * src/powerpc/linux64.S: Likewise.
3122 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3123 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3124 if -fpic/-fPIC/-mrelocatable.
3125 * src/powerpc/powerpc_closure.S: Likewise.
3126 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3127 #write in .eh_frame flags.
3128 * src/sparc/v9.S: Likewise.
3129 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3130 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3131 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3133 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3135 Fixes PR bootstrap/10656
3136 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3137 support for .register pseudo-op.
3138 * src/sparc/v8.S: Use it.
3139 * fficonfig.h.in: Rebuilt.
3140 * configure: Rebuilt.
3142 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3144 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3145 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3146 Make it the default on POWERPC64.
3147 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3148 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3149 * configure: Rebuilt.
3150 * src/powerpc/ffi.c (hidden): Define.
3151 (ffi_prep_args_SYSV): Renamed from
3152 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3153 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3154 ASM_NEEDS_REGISTERS64): New.
3155 (ffi_prep_args64): New function.
3156 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3157 (ffi_call): Likewise.
3158 (ffi_prep_closure): Likewise.
3159 (flush_icache): Surround by #ifndef POWERPC64.
3160 (ffi_dblfl): New union type.
3161 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3162 (ffi_closure_helper_LINUX64): New function.
3163 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3165 * src/powerpc/sysv.S: Likewise.
3166 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3167 * src/powerpc/linux64.S: New file.
3168 * src/powerpc/linux64_closure.S: New file.
3169 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3170 src/powerpc/linux64_closure.S.
3171 (TARGET_SRC_POWERPC): Likewise.
3173 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3174 closure_test_fn3): Fix result printing on big-endian 64-bit
3176 (main): Print tst2_arg instead of uninitialized tst2_result.
3178 * src/ffitest.c (main): Hide what closure pointer really points to
3181 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3183 * configure.in (arm-*-netbsdelf*): Add configuration.
3184 (configure): Regenerated.
3186 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3188 * include/Makefile.in: Regenerate.
3190 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3192 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3194 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3195 Receive closure pointer through parameter, read args using
3196 __builtin_dwarf_cfa.
3197 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3199 2003-03-12 Andreas Schwab <schwab@suse.de>
3201 * configure.in: Avoid trailing /. in toolexeclibdir.
3202 * configure: Rebuilt.
3204 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3206 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3208 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3210 * libffi/src/powerpc/darwin_closure.S:
3211 Fix alignement bug, allocate 8 bytes for the result.
3212 * libffi/src/powerpc/aix_closure.S:
3214 * libffi/src/powerpc/ffi_darwin.c:
3215 Update stackframe description for aix/darwin_closure.S.
3217 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3219 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3222 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3223 Andreas Schwab <schwab@suse.de>
3225 * configure.in: Adjust command to source config-ml.in to account
3226 for changes to the libffi_basedir definition.
3227 (libffi_basedir): Remove ${srcdir} from value and include trailing
3230 * configure: Regenerate.
3232 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3234 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3236 2003-01-28 Andrew Haley <aph@redhat.com>
3238 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3239 * src/x86/ffi64.c (ffi_prep_closure): New.
3240 (ffi_closure_UNIX64_inner): New.
3241 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3243 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3245 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3246 Remove USE_LIBDIR conditional.
3247 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3248 * Makefile.in, configure: Rebuilt.
3250 2003-01027 David Edelsohn <edelsohn@gnu.org>
3252 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3253 * Makefile.in: Regenerate.
3255 2003-01-22 Andrew Haley <aph@redhat.com>
3257 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3260 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3262 * src/powerpc/darwin.S: Add unwind info.
3263 * src/powerpc/darwin_closure.S: Likewise.
3265 2003-01-14 Andrew Haley <aph@redhat.com>
3267 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3268 (ffi_prep_cif_machdep): Likewise.
3269 * src/x86/unix64.S: Add unwind info.
3271 2003-01-14 Andreas Jaeger <aj@suse.de>
3273 * src/ffitest.c (main): Only use ffi_closures if those are
3276 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3278 * libffi/src/ffitest.c
3279 add closure testcases
3281 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3283 * libffi/src/powerpc/ffi.c
3284 fix alignment bug for float (4 byte aligned iso 8 byte)
3286 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3288 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3289 * src/powerpc/darwin.S: Remove RCS version string.
3291 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3293 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3294 * src/ffitest.c (main): Use static storage for closure.
3295 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3296 * src/sparc/v8.S (ffi_closure_v8): New.
3297 * src/sparc/v9.S (ffi_closure_v9): New.
3299 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3301 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3302 enumeration for X86_WIN32.
3303 * src/x86/win32.S: Added ffi_call_STDCALL function
3305 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3306 switch cases for recognising FFI_STDCALL and
3307 calling ffi_call_STDCALL if target is X86_WIN32.
3308 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3309 stdcall versions of the "my_strlen" and "many"
3310 test functions (for X86_WIN32).
3311 Added test cases to test stdcall invocation using
3314 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3316 * src/sh/sysv.S: Add DWARF2 unwind info.
3318 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3320 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3322 2002-11-26 Jim Wilson <wilson@redhat.com>
3324 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3326 2002-11-23 H.J. Lu <hjl@gnu.org>
3328 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3329 Include ../config/accross.m4.
3330 * aclocal.m4; Rebuild.
3331 * configure: Likewise.
3333 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3335 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3337 2002-11-11 DJ Delorie <dj@redhat.com>
3339 * configure.in: Look for common files in the right place.
3341 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3343 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3344 raw data as _Jv_word values, not ffi_raw.
3345 (ffi_java_ptrarray_to_raw): Likewise.
3346 (ffi_java_rvalue_to_raw): New function.
3347 (ffi_java_raw_call): Call it.
3348 (ffi_java_raw_to_rvalue): New function.
3349 (ffi_java_translate_args): Call it.
3350 * src/ffitest.c (closure_test_fn): Interpret return value
3351 as ffi_arg, not int.
3352 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3353 FFI_TYPE_POINTER case.
3354 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3355 values extended to word size.
3357 2002-10-02 Andreas Jaeger <aj@suse.de>
3359 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3361 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3363 * include/ffi.h.in: Fix i386 win32 compilation.
3365 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3367 * configure.in: Add s390x-*-linux-* target.
3368 * configure: Regenerate.
3369 * include/ffi.h.in: Define S390X for s390x targets.
3370 (FFI_CLOSURES): Define for s390/s390x.
3371 (FFI_TRAMPOLINE_SIZE): Likewise.
3372 (FFI_NATIVE_RAW_API): Likewise.
3373 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3374 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3375 * src/s390/ffi.c: Major rework of existing code. Add support for
3376 s390x targets. Add closure support.
3377 * src/s390/sysv.S: Likewise.
3379 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3381 * src/arm/sysv.S: Fix typo.
3383 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3385 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3386 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3387 (ffi_call_SYSV): Handle soft-float.
3389 2002-09-27 Bo Thorsen <bo@suse.de>
3391 * include/ffi.h.in: Fix multilib x86-64 support.
3393 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3395 * Makefile.am (all-multi): Fix multilib parallel build.
3397 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3399 * configure.in (sh[34]*-*-linux*): Add brackets.
3400 * configure: Regenerate.
3402 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3404 * Makefile.am: Add SH support.
3405 * Makefile.in: Regenerate.
3406 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3407 * configure: Regenerate.
3408 * include/ffi.h.in: Add SH support.
3409 * src/sh/ffi.c: New file.
3410 * src/sh/sysv.S: New file.
3411 * src/types.c: Add SH support.
3413 2002-07-16 Bo Thorsen <bo@suse.de>
3415 * src/x86/ffi64.c: New file that adds x86-64 support.
3416 * src/x86/unix64.S: New file that handles argument setup for
3418 * src/x86/sysv.S: Don't use this on x86-64.
3419 * src/x86/ffi.c: Don't use this on x86-64.
3421 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3423 * src/ffitest.c (struct6): New test that tests a special case in
3425 (struct7): Likewise.
3426 (struct8): Likewise.
3427 (struct9): Likewise.
3428 (closure_test_fn): Silence warning about this when it's not used.
3429 (main): Add the new tests.
3430 (main): Fix a couple of wrong casts and silence some compiler warnings.
3431 * include/ffi.h.in: Add x86-64 ABI definition.
3432 * fficonfig.h.in: Regenerate.
3433 * Makefile.am: Add x86-64 support.
3434 * configure.in: Likewise.
3435 * Makefile.in: Regenerate.
3436 * configure: Likewise.
3438 2002-06-24 Bo Thorsen <bo@suse.de>
3440 * src/types.c: Merge settings for similar architectures.
3441 Add x86-64 sizes and alignments.
3443 2002-06-23 Bo Thorsen <bo@suse.de>
3445 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3446 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3447 * src/mips/ffi.c (ffi_prep_args): Likewise.
3448 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3450 2002-07-18 H.J. Lu (hjl@gnu.org)
3452 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3453 (libffi_la_SOURCES): Support MIPS_LINUX.
3454 (libffi_convenience_la_SOURCES): Likewise.
3455 * Makefile.in: Regenerated.
3457 * configure.in (mips64*-*): Skip.
3458 (mips*-*-linux*): New.
3459 * configure: Regenerated.
3461 * src/mips/ffi.c: Include <sgidefs.h>.
3463 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3465 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3467 2002-05-27 Roger Sayle <roger@eyesopen.com>
3469 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3471 2002-05-27 Bo Thorsen <bo@suse.de>
3473 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3476 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3478 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3479 beginning of function (for older apple cc).
3481 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3483 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3484 script entry, and set LD to it when configuring multilibs.
3485 * configure: Rebuilt.
3487 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3489 * configure.in (sparc64-*-netbsd*): Add target.
3490 (sparc-*-netbsdelf*): Likewise.
3491 * configure: Regenerate.
3493 2002-04-28 David S. Miller <davem@redhat.com>
3495 * configure.in, configure: Fix SPARC test in previous change.
3497 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3499 * Makefile.am: Add Linux for S/390 support.
3500 * Makefile.in: Regenerate.
3501 * configure.in: Add Linux for S/390 support.
3502 * configure: Regenerate.
3503 * include/ffi.h.in: Add Linux for S/390 support.
3504 * src/s390/ffi.c: New file from libffi CVS tree.
3505 * src/s390/sysv.S: New file from libffi CVS tree.
3507 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3509 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3511 * src/sparc/v8.S: Use it.
3512 * src/sparc/v9.S: Likewise.
3513 * fficonfig.h.in: Rebuilt.
3514 * configure: Rebuilt.
3516 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3518 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3520 * src/ia64/unix.S: Add unwind information. Fix comments.
3521 Save sp in a way that's compatible with unwind info.
3522 (ffi_call_unix): Correctly restore sp in all cases.
3523 * src/ia64/ffi.c: Add, fix comments.
3525 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3527 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3529 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3531 * configure.in (alpha*-*-netbsd*): Add target.
3532 * configure: Regenerate.
3534 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3536 * src/sparc/v8.S: Add unwind info.
3537 * src/sparc/v9.S: Likewise.
3539 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3541 * configure.in: Enable i*86-*-netbsdelf*.
3542 * configure: Rebuilt.
3544 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3547 * src/mips/n32.s: Delete
3548 * src/mips/o32.s: Delete
3550 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3552 * configure.in: Enable alpha*-*-freebsd*.
3553 * configure: Rebuilt.
3555 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3557 * Makefile.am: libfficonvenience -> libffi_convenience.
3558 * Makefile.in: Rebuilt.
3560 * Makefile.am: Define ffitest_OBJECTS.
3561 * Makefile.in: Rebuilt.
3563 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3564 David Edelsohn <edelsohn@gnu.org>
3566 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3567 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3568 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3569 * Makefile.in: Regenerate.
3570 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3571 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3572 (flush_icache, flush_range): New functions.
3573 (ffi_closure_helper_DARWIN): New function.
3574 * src/powerpc/aix_closure.S: New file.
3575 * src/powerpc/darwin_closure.S: New file.
3577 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3579 * include/ffi.h.in: Add typedef for ffi_arg.
3580 * src/ffitest.c (main): Declare rint with ffi_arg.
3582 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3584 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3585 number of GPRs for floating-point arguments.
3587 2002-01-31 Anthony Green <green@redhat.com>
3589 * configure: Rebuilt.
3590 * configure.in: Replace CHECK_SIZEOF and endian tests with
3591 cross-compiler friendly macros.
3592 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3595 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3597 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3598 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3600 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3602 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3603 (TARGET_SRC_POWERPC_AIX): New.
3604 (POWERPC_AIX): New stanza.
3605 * Makefile.in: Regenerate.
3606 * configure.in: Add AIX case.
3607 * configure: Regenerate.
3608 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3609 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3610 size. Fix "long double" support.
3611 (ffi_call): Add FFI_AIX case.
3612 * src/powerpc/aix.S: New.
3614 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3616 Implement Darwin PowerPC ABI.
3617 * configure.in: Handle powerpc-*-darwin*.
3618 * Makefile.am: Set source files for POWERPC_DARWIN.
3619 * configure: Rebuilt.
3620 * Makefile.in: Rebuilt.
3621 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3623 * src/powerpc/darwin.S: New file.
3624 * src/powerpc/ffi_darwin.c: New file.
3626 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3628 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3630 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3632 * src/x86/sysv.S: Avoid gas-only .balign directive.
3633 Use C style comments.
3635 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3637 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3638 Fixes PR bootstrap/3563.
3640 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3642 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3644 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3646 * configure.in: Recognize sparc*-sun-* host.
3647 * configure: Regenerate.
3649 2001-06-06 Andrew Haley <aph@redhat.com>
3651 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3653 2001-06-03 Andrew Haley <aph@redhat.com>
3655 * src/alpha/osf.S: Add unwind info.
3656 * src/powerpc/sysv.S: Add unwind info.
3657 * src/powerpc/ppc_closure.S: Likewise.
3659 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3661 * configure.in: Fix AC_ARG_ENABLE usage.
3662 * configure: Rebuilt.
3664 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3666 * configure.in: Remove warning about beta code.
3667 * configure: Rebuilt.
3669 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3671 * src/ia64/unix.S: Restore stack pointer when returning from
3673 * src/ia64/ffi.c: Fix typo in comment.
3675 2001-04-18 Jim Wilson <wilson@redhat.com>
3677 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3678 to eliminate RAW DV.
3680 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3682 * Makefile.am: Make a libtool convenience library.
3683 * Makefile.in: Rebuilt.
3685 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3687 * configure.in: Use different syntax for subdirectory creation.
3688 * configure: Rebuilt.
3690 2001-03-27 Jon Beniston <jon@beniston.com>
3692 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3693 * configure: Rebuilt.
3694 * Makefile.am: Added X86_WIN32 target support.
3695 * Makefile.in: Rebuilt.
3697 * include/ffi.h.in: Added X86_WIN32 target support.
3699 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3700 * src/types.c: Added X86_WIN32 target support.
3702 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3703 stuff removed and made to work with CygWin's gas.
3705 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3707 * configure.in: Make target subdirectory in build dir.
3708 * Makefile.am: Override suffix based rules to specify correct output
3710 * Makefile.in: Rebuilt.
3711 * configure: Rebuilt.
3713 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3715 * src/powerpc/ppc_closure.S: New file.
3716 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3717 involving long long and register pairs.
3718 (ffi_prep_closure): New function.
3719 (flush_icache): Likewise.
3720 (ffi_closure_helper_SYSV): Likewise.
3721 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3722 (FFI_TRAMPOLINE_SIZE): Likewise.
3723 (FFI_NATIVE_RAW_API): Likewise.
3724 * Makefile.in: Rebuilt.
3725 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3726 (TARGET_SRC_POWERPC): Likewise.
3728 2001-03-19 Tom Tromey <tromey@redhat.com>
3730 * Makefile.in: Rebuilt.
3731 * Makefile.am (ffitest_LDFLAGS): New macro.
3733 2001-03-02 Nick Clifton <nickc@redhat.com>
3735 * include/ffi.h.in: Remove RCS ident string.
3736 * include/ffi_mips.h: Remove RCS ident string.
3737 * src/debug.c: Remove RCS ident string.
3738 * src/ffitest.c: Remove RCS ident string.
3739 * src/prep_cif.c: Remove RCS ident string.
3740 * src/types.c: Remove RCS ident string.
3741 * src/alpha/ffi.c: Remove RCS ident string.
3742 * src/alpha/osf.S: Remove RCS ident string.
3743 * src/arm/ffi.c: Remove RCS ident string.
3744 * src/arm/sysv.S: Remove RCS ident string.
3745 * src/mips/ffi.c: Remove RCS ident string.
3746 * src/mips/n32.S: Remove RCS ident string.
3747 * src/mips/o32.S: Remove RCS ident string.
3748 * src/sparc/ffi.c: Remove RCS ident string.
3749 * src/sparc/v8.S: Remove RCS ident string.
3750 * src/sparc/v9.S: Remove RCS ident string.
3751 * src/x86/ffi.c: Remove RCS ident string.
3752 * src/x86/sysv.S: Remove RCS ident string.
3754 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3756 * include/ffi.h.in: Change sourceware.cygnus.com references to
3759 2000-12-09 Richard Henderson <rth@redhat.com>
3761 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3762 (ffi_closure_osf_inner): Index rather than increment avalue
3763 and arg_types. Give ffi_closure_osf the raw return value type.
3764 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3767 2000-12-07 Richard Henderson <rth@redhat.com>
3769 * src/raw_api.c (ffi_translate_args): Fix typo.
3770 (ffi_prep_closure): Likewise.
3772 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3773 FFI_TRAMPOLINE_SIZE.
3774 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3775 cif->bytes for new ffi_call_osf implementation.
3776 (ffi_prep_args): Absorb into ...
3777 (ffi_call): ... here. Do all stack allocation here and
3778 avoid a callback function.
3779 (ffi_prep_closure, ffi_closure_osf_inner): New.
3780 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3781 (ffi_closure_osf): New.
3783 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3785 * config.guess, config.sub, install-sh: Removed.
3786 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3787 * Makefile.in: Rebuilt.
3789 * acinclude.m4: Include libtool macros from the top level.
3790 * aclocal.m4, configure: Rebuilt.
3792 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3794 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3795 * configure: Rebuilt.
3797 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3799 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3800 memory correctly. Use conditional instructions, not branches where
3803 2000-05-04 Tom Tromey <tromey@cygnus.com>
3805 * configure: Rebuilt.
3806 * configure.in: Match `arm*-*-linux-*'.
3807 From Chris Dornan <cdornan@arm.com>.
3809 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3811 * Makefile.am (SUBDIRS): Define.
3812 (AM_MAKEFLAGS): Likewise.
3813 (Multilib support.): Add section.
3814 * Makefile.in: Rebuilt.
3815 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3816 New variables. Set for gcc using -print-multi-lib. Export them
3818 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3819 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3821 (extra_compiler_flags_value, extra_compiler_flags): Check these
3822 for extra compiler options which need to be passed down in
3825 2000-04-16 Anthony Green <green@redhat.com>
3827 * configure: Rebuilt.
3828 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3830 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3832 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3833 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3834 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3835 Replace all void * sizeofs with sizeof(int).
3836 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3837 different than DOUBLE.
3838 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3839 (ffi_prep_args_v9): New function.
3840 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3841 (ffi_V9_return_struct): New function.
3842 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3843 32bit code (not yet cross-arch calls).
3844 * src/sparc/v8.S: Add struct return delay nop.
3846 * src/sparc/v9.S: New file.
3847 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3848 is used on sparc64 only for structures larger than 32 bytes.
3849 Pass by reference for structures is done for structure arguments
3850 larger than 16 bytes.
3851 * src/ffitest.c (main): Use 64bit rint on sparc64.
3852 Run long long tests on sparc.
3853 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3855 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3857 * configure.in (sparc-*-linux*): New supported target.
3858 (sparc64-*-linux*): Likewise.
3859 * configure: Rebuilt.
3860 * Makefile.am: Add v9.S to SPARC files.
3861 * Makefile.in: Likewise.
3862 (LINK): Surround $(CCLD) into double quotes, so that multilib
3863 compiles work correctly.
3865 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3867 * configure: Rebuilt.
3868 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3869 proposed by Bryce McKinlay.
3871 2000-03-20 Tom Tromey <tromey@cygnus.com>
3873 * Makefile.in: Hand edit for java_raw_api.lo.
3875 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3877 * config.guess, config.sub: Update from the gcc tree.
3878 Fix for PR libgcj/168.
3880 2000-03-03 Tom Tromey <tromey@cygnus.com>
3882 * Makefile.in: Fixed ia64 by hand.
3884 * configure: Rebuilt.
3885 * configure.in (--enable-multilib): New option.
3886 (libffi_basedir): New subst.
3887 (AC_OUTPUT): Added multilib code.
3889 2000-03-02 Tom Tromey <tromey@cygnus.com>
3891 * Makefile.in: Rebuilt.
3892 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3895 2000-02-25 Hans Boehm <boehm@acm.org>
3897 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3899 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3901 (ffi_prep_raw_closure): Use ffi_translate_args, not
3902 ffi_closure_translate.
3903 * src/java_raw_api.c: New file.
3904 * src/ffitest.c (closure_test_fn): New function.
3905 (main): Define `rint' as long long on IA64. Added new test when
3906 FFI_CLOSURES is defined.
3907 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3908 (ffi_abi): Recognize IA64.
3909 (ffi_raw): Added `flt' field.
3910 Added "Java raw API" code.
3911 * configure.in: Recognize ia64.
3912 * Makefile.am (TARGET_SRC_IA64): New macro.
3913 (libffi_la_common_SOURCES): Added java_raw_api.c.
3914 (libffi_la_SOURCES): Define in IA64 case.
3916 2000-01-04 Tom Tromey <tromey@cygnus.com>
3918 * Makefile.in: Rebuilt with newer automake.
3920 1999-12-31 Tom Tromey <tromey@cygnus.com>
3922 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3924 1999-09-01 Tom Tromey <tromey@cygnus.com>
3926 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3928 * fficonfig.h.in: Rebuilt.
3929 * configure: Rebuilt.
3930 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3931 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3932 * acconfig.h: Don't #undef PACKAGE or VERSION.
3934 1999-08-09 Anthony Green <green@cygnus.com>
3936 * include/ffi.h.in: Try to work around messy header problem
3937 with PACKAGE and VERSION.
3939 * configure: Rebuilt.
3940 * configure.in: Change version to 2.00-beta.
3942 * fficonfig.h.in: Rebuilt.
3943 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3945 * src/x86/ffi.c (ffi_raw_call): Rename.
3947 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3949 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3950 (ffi_prep_incoming_args_SYSV): Ditto.
3951 (ffi_prep_closure): Ditto.
3952 (ffi_closure_raw_SYSV): Ditto.
3953 (ffi_prep_raw_closure): More ditto.
3954 (ffi_call_raw): Final ditto.
3956 * include/ffi.h.in: Add definitions for closure and raw API.
3958 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3961 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3963 * src/raw_api.c: New file.
3965 * include/ffi.h.in (ffi_raw): New type.
3966 (UINT_ARG, SINT_ARG): New defines.
3967 (ffi_closure, ffi_raw_closure): New types.
3968 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3970 * configure.in: Add check for endianness and sizeof void*.
3972 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3973 instead of directly.
3975 * configure: Rebuilt.
3977 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3979 * configure.in: Add x86 and powerpc BeOS configurations.
3980 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3982 1999-05-09 Anthony Green <green@cygnus.com>
3984 * configure.in: Add warning about this being beta code.
3985 Remove src/Makefile.am from the picture.
3986 * configure: Rebuilt.
3988 * Makefile.am: Move logic from src/Makefile.am. Add changes
3989 to support libffi as a target library.
3990 * Makefile.in: Rebuilt.
3992 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3993 Upgraded to new autoconf, automake, libtool.
3997 * LICENSE: Update copyright date.
3999 * src/Makefile.am, src/Makefile.in: Removed.
4001 1998-11-29 Anthony Green <green@cygnus.com>
4003 * include/ChangeLog: Removed.
4004 * src/ChangeLog: Removed.
4005 * src/mips/ChangeLog: Removed.
4006 * src/sparc/ChangeLog: Remboved.
4007 * src/x86/ChangeLog: Removed.
4009 * ChangeLog.v1: Created.